id
string | submitter
string | authors
string | title
string | comments
string | journal-ref
string | doi
string | report-no
string | categories
string | license
string | abstract
string | versions
list | update_date
timestamp[s] | authors_parsed
list |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2505.03567
|
Zengli Luo
|
Zengli Luo, Canlong Zhang, Zhixin Li, Zhiwen Wang, Chunrong Wei
|
Uncertainty-Aware Prototype Semantic Decoupling for Text-Based Person Search in Full Images
|
9 pages, 5 figures. Accepted by the 18th International Conference on Knowledge Science, Engineering and Management (KSEM 2025)
| null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Text-based pedestrian search (TBPS) in full images aims to locate a target pedestrian in untrimmed images using natural language descriptions. However, in complex scenes with multiple pedestrians, existing methods are limited by uncertainties in detection and matching, leading to degraded performance. To address this, we propose UPD-TBPS, a novel framework comprising three modules: Multi-granularity Uncertainty Estimation (MUE), Prototype-based Uncertainty Decoupling (PUD), and Cross-modal Re-identification (ReID). MUE conducts multi-granularity queries to identify potential targets and assigns confidence scores to reduce early-stage uncertainty. PUD leverages visual context decoupling and prototype mining to extract features of the target pedestrian described in the query. It separates and learns pedestrian prototype representations at both the coarse-grained cluster level and the fine-grained individual level, thereby reducing matching uncertainty. ReID evaluates candidates with varying confidence levels, improving detection and retrieval accuracy. Experiments on CUHK-SYSU-TBPS and PRW-TBPS datasets validate the effectiveness of our framework.
|
[
{
"version": "v1",
"created": "Tue, 06 May 2025 14:25:30 GMT"
},
{
"version": "v2",
"created": "Wed, 07 May 2025 01:21:36 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 03:06:50 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Luo",
"Zengli",
""
],
[
"Zhang",
"Canlong",
""
],
[
"Li",
"Zhixin",
""
],
[
"Wang",
"Zhiwen",
""
],
[
"Wei",
"Chunrong",
""
]
] |
2505.03735
|
Haoning Wu
|
Jiayuan Rao, Zifeng Li, Haoning Wu, Ya Zhang, Yanfeng Wang, Weidi Xie
|
Multi-Agent System for Comprehensive Soccer Understanding
|
Accepted by ACM MM 2025; Project Page: https://jyrao.github.io/SoccerAgent/
| null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Recent advances in soccer understanding have demonstrated rapid progress, yet existing research predominantly focuses on isolated or narrow tasks. To bridge this gap, we propose a comprehensive framework for holistic soccer understanding. Concretely, we make the following contributions in this paper: (i) we construct SoccerWiki, the first large-scale multimodal soccer knowledge base, integrating rich domain knowledge about players, teams, referees, and venues to enable knowledge-driven reasoning; (ii) we present SoccerBench, the largest and most comprehensive soccer-specific benchmark, featuring around 10K multimodal (text, image, video) multi-choice QA pairs across 13 distinct tasks; (iii) we introduce SoccerAgent, a novel multi-agent system that decomposes complex soccer questions via collaborative reasoning, leveraging domain expertise from SoccerWiki and achieving robust performance; (iv) extensive evaluations and comparisons with representative MLLMs on SoccerBench highlight the superiority of our agentic system.
|
[
{
"version": "v1",
"created": "Tue, 06 May 2025 17:59:31 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 03:17:55 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Rao",
"Jiayuan",
""
],
[
"Li",
"Zifeng",
""
],
[
"Wu",
"Haoning",
""
],
[
"Zhang",
"Ya",
""
],
[
"Wang",
"Yanfeng",
""
],
[
"Xie",
"Weidi",
""
]
] |
2505.04085
|
Minseok Kim
|
Minseok Kim, Gesi Teng, Keita Nishi, Togo Ikegami, Masamune Sato
|
Device-Free Localization Using Multi-Link MIMO Channels in Distributed Antenna Networks
| null | null |
10.1109/JSTEAP.2025.3603541
| null |
eess.SP eess.IV
|
http://creativecommons.org/licenses/by/4.0/
|
Targeting integrated sensing and communication (ISAC) in future 6G radio access networks (RANs), this paper presents a novel device-free localization (DFL) framework based on distributed antenna networks (DANs). In the proposed approach, radio tomographic imaging (RTI) leverages the spatial and temporal diversity of multi-link multiple-input multiple-output (MIMO) channels in DANs to achieve accurate localization. Furthermore, a prototype system was developed using software-defined radios (SDRs) operating in the sub-6 GHz band, and comprehensive evaluations were conducted under indoor conditions involving varying node densities and target types. The results demonstrate that the framework achieves sub-meter localization accuracy in most scenarios and maintains robust performance under complex multipath environments. In addition, the use of Bayesian optimization to fine-tune key parameters, such as sparsity and path thickness, led to significant improvements in image reconstruction quality and target estimation accuracy. These results demonstrate the feasibility and effectiveness of DAN-based DFL as a scalable and infrastructure-compatible ISAC solution, capable of delivering accurate, passive localization without dedicated sensing hardware.
|
[
{
"version": "v1",
"created": "Wed, 07 May 2025 02:54:00 GMT"
},
{
"version": "v2",
"created": "Mon, 21 Jul 2025 14:19:08 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Kim",
"Minseok",
""
],
[
"Teng",
"Gesi",
""
],
[
"Nishi",
"Keita",
""
],
[
"Ikegami",
"Togo",
""
],
[
"Sato",
"Masamune",
""
]
] |
2505.04138
|
Hongyu Lu
|
Hongyu Lu, Han-Qing Wu, Bin-Bin Chen, Wang Yao, and Zi Yang Meng
|
Generic (fractional) quantum anomalous Hall crystals from interaction-driven band folding
|
4 figures in the updated main text
| null | null | null |
cond-mat.str-el cond-mat.mes-hall
|
http://creativecommons.org/licenses/by/4.0/
|
Among the extensive studies of fractional quantum anomalous Hall (FQAH) states, there recently appears a growing interest in the topological states with coexisting charge density wave (CDW) orders. Such states are referred to as Hall crystals. However, compared to those with integer Hall conductivities, the FQAH crystal (FQAHC) is still elusive even at the level of microscopic model. In this work, we numerically study a topological flat-band model on triangular lattice with spinless fermions. At fractional filling of the Chern band, the nearest-neighbor interaction leads to a commensurate and topologically trivial CDW state. Interestingly, the folded mini-band above the CDW gap is non-trivial, and we focus on the doping of it without any projection. A series of (F)QAHC states at (fractional) integer fillings of this mini-band are discovered and some FQAHC state might even exist in less "ideal" conditions. The ground-state degeneracies of such (F)QAHC states are enlarged by the CDW degeneracy and the Hall conductivities -- determined by the fillings of the mini-band -- are different from the fillings of the original Chern band. We also study the thermodynamics of an FQAHC state and find a compressible CDW phase at intermediate temperatures, which might serve as a precursor of lower temperature FQAHC phase. Moreover, we numerically demonstrate that such a generic scheme of doping CDW-folded topological mini-band could be applied to bosonic systems, broadening the platforms of Hall-crystal physics and motivating its exploration in quantum moire and cold-atom systems.
|
[
{
"version": "v1",
"created": "Wed, 07 May 2025 05:25:02 GMT"
},
{
"version": "v2",
"created": "Fri, 23 May 2025 13:38:50 GMT"
},
{
"version": "v3",
"created": "Sun, 31 Aug 2025 04:19:08 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Lu",
"Hongyu",
""
],
[
"Wu",
"Han-Qing",
""
],
[
"Chen",
"Bin-Bin",
""
],
[
"Yao",
"Wang",
""
],
[
"Meng",
"Zi Yang",
""
]
] |
2505.04615
|
Md Riajul Haque
|
Md Riajul Haque, Sourav Pal and Debarun Paul
|
Improved Predictions on Higgs-Starobinsky Inflation and Reheating with ACT DR6 and Primordial Gravitational Waves
|
Accepted for publication in PLB, 7 pages, 2 figures, 2 tables
| null |
10.1016/j.physletb.2025.139852
| null |
astro-ph.CO hep-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We investigate the implications of recent CMB observations for Higgs-Starobinsky inflationary models and their associated reheating dynamics, utilizing data from ACT DR6, Planck 2018, BICEP/Keck 2018, and DESI, collectively referred to as P-ACT-LB-BK18. In addition to direct CMB constraints, we incorporate indirect bounds arising from the potential overproduction of primordial gravitational waves (PGWs), particularly through limits on the effective number of relativistic species, $\Delta N_{\rm eff}$, during Big Bang Nucleosynthesis (BBN). These constraints become especially relevant in scenarios featuring a stiff post-inflationary equation of state $w_{\rm RH}\geq 0.58$. Our analysis shows that, when both P-ACT-LB-BK18 data and $\Delta N_{\rm eff}$ bounds are considered, the viable number of inflationary e-folds is restricted to the range ($57.9$-$62.2$) at the $2\sigma$ confidence level (C.L.). Correspondingly, the reheating temperature is constrained to lie between the BBN energy scale and $10^{12}$ GeV, with the post-inflationary equation-of-state parameter satisfying $w_{\rm RH} > 0.41$. However, no parameter space remains viable at the $1\sigma$ C.L. once $\Delta N_{\rm eff}$ constraints from PGWs are included, rendering the Higgs-Starobinsky model highly restricted.
|
[
{
"version": "v1",
"created": "Wed, 07 May 2025 17:58:13 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 15:47:16 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Haque",
"Md Riajul",
""
],
[
"Pal",
"Sourav",
""
],
[
"Paul",
"Debarun",
""
]
] |
2505.04802
|
Xiao Wang
|
Xiao Wang, Jong-Youl Choi, Takuya Kurihaya, Isaac Lyngaas, Hong-Jun Yoon, Xi Xiao, David Pugmire, Ming Fan, Nasik M. Nafi, Aristeidis Tsaris, Ashwin M. Aji, Maliha Hossain, Mohamed Wahib, Dali Wang, Peter Thornton, Prasanna Balaprakash, Moetasim Ashfaq, Dan Lu
|
ORBIT-2: Scaling Exascale Vision Foundation Models for Weather and Climate Downscaling
| null | null | null | null |
cs.LG astro-ph.EP cs.AI cs.DC physics.ao-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Sparse observations and coarse-resolution climate models limit effective regional decision-making, underscoring the need for robust downscaling. However, existing AI methods struggle with generalization across variables and geographies and are constrained by the quadratic complexity of Vision Transformer (ViT) self-attention. We introduce ORBIT-2, a scalable foundation model for global, hyper-resolution climate downscaling. ORBIT-2 incorporates two key innovations: (1) Residual Slim ViT (Reslim), a lightweight architecture with residual learning and Bayesian regularization for efficient, robust prediction; and (2) TILES, a tile-wise sequence scaling algorithm that reduces self-attention complexity from quadratic to linear, enabling long-sequence processing and massive parallelism. ORBIT-2 scales to 10 billion parameters across 65,536 GPUs, achieving up to 4.1 exaFLOPS sustained throughput and 74--98% strong scaling efficiency. It supports downscaling to 0.9 km global resolution and processes sequences up to 4.2 billion tokens. On 7 km resolution benchmarks, ORBIT-2 achieves high accuracy with $R^2$ scores in the range of 0.98--0.99 against observational data.
|
[
{
"version": "v1",
"created": "Wed, 07 May 2025 21:09:00 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 03:08:37 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Wang",
"Xiao",
""
],
[
"Choi",
"Jong-Youl",
""
],
[
"Kurihaya",
"Takuya",
""
],
[
"Lyngaas",
"Isaac",
""
],
[
"Yoon",
"Hong-Jun",
""
],
[
"Xiao",
"Xi",
""
],
[
"Pugmire",
"David",
""
],
[
"Fan",
"Ming",
""
],
[
"Nafi",
"Nasik M.",
""
],
[
"Tsaris",
"Aristeidis",
""
],
[
"Aji",
"Ashwin M.",
""
],
[
"Hossain",
"Maliha",
""
],
[
"Wahib",
"Mohamed",
""
],
[
"Wang",
"Dali",
""
],
[
"Thornton",
"Peter",
""
],
[
"Balaprakash",
"Prasanna",
""
],
[
"Ashfaq",
"Moetasim",
""
],
[
"Lu",
"Dan",
""
]
] |
2505.05287
|
Li Zechu
|
Zechu Li, Yufeng Jin, Daniel Ordonez Apraez, Claudio Semini, Puze Liu, Georgia Chalvatzaki
|
Morphologically Symmetric Reinforcement Learning for Ambidextrous Bimanual Manipulation
| null | null | null | null |
cs.RO cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Humans naturally exhibit bilateral symmetry in their gross manipulation skills, effortlessly mirroring simple actions between left and right hands. Bimanual robots-which also feature bilateral symmetry-should similarly exploit this property to perform tasks with either hand. Unlike humans, who often favor a dominant hand for fine dexterous skills, robots should ideally execute ambidextrous manipulation with equal proficiency. To this end, we introduce SYMDEX (SYMmetric DEXterity), a reinforcement learning framework for ambidextrous bi-manipulation that leverages the robot's inherent bilateral symmetry as an inductive bias. SYMDEX decomposes complex bimanual manipulation tasks into per-hand subtasks and trains dedicated policies for each. By exploiting bilateral symmetry via equivariant neural networks, experience from one arm is inherently leveraged by the opposite arm. We then distill the subtask policies into a global ambidextrous policy that is independent of the hand-task assignment. We evaluate SYMDEX on six challenging simulated manipulation tasks and demonstrate successful real-world deployment on two of them. Our approach strongly outperforms baselines on complex task in which the left and right hands perform different roles. We further demonstrate SYMDEX's scalability by extending it to a four-arm manipulation setup, where our symmetry-aware policies enable effective multi-arm collaboration and coordination. Our results highlight how structural symmetry as inductive bias in policy learning enhances sample efficiency, robustness, and generalization across diverse dexterous manipulation tasks.
|
[
{
"version": "v1",
"created": "Thu, 08 May 2025 14:29:00 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 18:49:59 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Li",
"Zechu",
""
],
[
"Jin",
"Yufeng",
""
],
[
"Apraez",
"Daniel Ordonez",
""
],
[
"Semini",
"Claudio",
""
],
[
"Liu",
"Puze",
""
],
[
"Chalvatzaki",
"Georgia",
""
]
] |
2505.05304
|
Rishita Das
|
Shreyashri Sarkar and Rishita Das
|
Information-theoretic characterization of turbulence intermittency
|
7 pages, 5 figures
| null | null | null |
physics.flu-dyn physics.data-an
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present an information-theoretic characterization of small-scale intermittency in turbulent flows, that distinguishes turbulence-induced intermittency from purely kinematic effects. Kullback-Leibler (KL) divergence is used to quantify the deviation of pseudodissipation, dissipation or enstrophy of a turbulent flow field from that of a Gaussian random velocity field, serving as a comprehensive measure of the small-scale intermittency arising from turbulence dynamics. Shannon entropy is employed to evaluate the uncertainty of these small-scale quantities. Analysis of direct numerical simulation data of forced homogeneous isotropic turbulent flow across a wide range of Taylor Reynolds numbers demonstrates the existence of two critical Reynolds numbers where the variation of uncertainty changes. The study reveals a new scaling behavior and presence of a symmetry: (i) turbulence-induced intermittency grows logarithmically with Reynolds number unlike the commonly reported power-law scaling and (ii) turbulence dynamics produce equal intermittency in both dissipation rate and enstrophy, in contrast to the prevailing assumptions of asymmetry in strain-rate and vorticity dynamics.
|
[
{
"version": "v1",
"created": "Thu, 08 May 2025 14:50:36 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 18:07:26 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Sarkar",
"Shreyashri",
""
],
[
"Das",
"Rishita",
""
]
] |
2505.05818
|
Vladimir Folomeev
|
Vladimir Dzhunushaliev, Vladimir Folomeev, and Sayabek Sakhiyev
|
Dirac stars with wormhole topology
|
18 pages, 10 figures, minor corrections to content, version published in GRG
|
Gen. Relativ. Gravit. 57, 126 (2025)
|
10.1007/s10714-025-03463-5
| null |
gr-qc
|
http://creativecommons.org/licenses/by/4.0/
|
We consider configurations consisting of a gravitating nonlinear spinor field and a massless ghost scalar field providing a nontrivial spacetime topology. For such a mixed system, we have constructed families of asymptotically flat asymmetric solutions describing Dirac stars with wormhole topology. The physical properties of such systems are completely determined by the values of three input quantities: the nonlinearity parameter of the spinor field, the spinor frequency, and the throat parameter. Depending on the specific values of these parameters, the configurations may be regular or singular and possess one or two throats, as well as they may contain possible horizons of different kinds. Furthermore, because of the asymmetry, masses and sizes of the configurations observed at the two asymptotic ends of the wormhole may differ considerably. For large negative values of the nonlinearity parameter, it is possible to obtain configurations whose masses are comparable to the Chandrasekhar mass and effective radii are of the order of kilometers.
|
[
{
"version": "v1",
"created": "Fri, 09 May 2025 06:43:20 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 05:26:18 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Dzhunushaliev",
"Vladimir",
""
],
[
"Folomeev",
"Vladimir",
""
],
[
"Sakhiyev",
"Sayabek",
""
]
] |
2505.05847
|
Jens Zentgraf
|
Johanna Elena Schmitz, Jens Zentgraf, Sven Rahmann
|
Smaller and More Flexible Cuckoo Filters
| null | null | null | null |
cs.DS cs.DB
|
http://creativecommons.org/licenses/by/4.0/
|
Cuckoo filters are space-efficient approximate set membership data structures with a controllable false positive rate (FPR) and zero false negatives, similar to Bloom filters. In contrast to Bloom filters, Cuckoo filters store multi-bit fingerprints of keys in a hash table using variants of Cuckoo hashing, allowing each fingerprint to be stored at a small number of possible locations. Existing Cuckoo filters use fingerprints of $(k+3)$ bits per key and an additional space overhead factor of at least $1.05$ to achieve an FPR of $2^{-k}$. For $k=10$, this amounts to $1.365\, kn$ bits to store $n$ keys, which is better than $1.443\, kn$ bits for Bloom filters. The $+3$ for the fingerprint size is required to balance out the multiplied FPR caused by looking for the fingerprint at several locations. In the original Cuckoo filter, the number of hash table buckets is restricted to a power of 2, which may lead to much larger space overheads, up to $2.1\, (1+3/k)\, kn$ bits.
We present two improvements of Cuckoo filters. First, we remove the restriction that the number of buckets must be a power of 2 by using a different placement strategy. Second, we reduce the space overhead factor of Cuckoo filters to $1.06 \, (1+2/k)$ by using overlapping windows instead of disjoint buckets to maintain the load threshold of the hash table, while reducing the number of alternative slots where any fingerprint may be found.
A detailed evaluation demonstrates that the alternative memory layout based on overlapping windows decreases the size of Cuckoo filters not only in theory, but also in practice. A comparison with other state-of-the art filter types, Prefix filters and Vector Quotient filters (VQFs), shows that the reduced space overhead makes windowed Cuckoo filters the smallest filters supporting online insertions, with similarly fast queries, but longer insertion times.
|
[
{
"version": "v1",
"created": "Fri, 09 May 2025 07:37:38 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 06:11:39 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Schmitz",
"Johanna Elena",
""
],
[
"Zentgraf",
"Jens",
""
],
[
"Rahmann",
"Sven",
""
]
] |
2505.05857
|
Nathan Justin
|
Nathan Justin, Qingshi Sun, Andr\'es G\'omez, Phebe Vayanos
|
Responsible Machine Learning via Mixed-Integer Optimization
|
69 pages, 12 figures
| null | null | null |
cs.LG math.OC stat.ML
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In the last few decades, Machine Learning (ML) has achieved significant success across domains ranging from healthcare, sustainability, and the social sciences, to criminal justice and finance. But its deployment in increasingly sophisticated, critical, and sensitive areas affecting individuals, the groups they belong to, and society as a whole raises critical concerns around fairness, transparency and robustness, among others. As the complexity and scale of ML systems and of the settings in which they are deployed grow, so does the need for responsible ML methods that address these challenges while providing guaranteed performance in deployment.
Mixed-integer optimization (MIO) offers a powerful framework for embedding responsible ML considerations directly into the learning process while maintaining performance. For example, it enables learning of inherently transparent models that can conveniently incorporate fairness or other domain specific constraints. This tutorial paper provides an accessible and comprehensive introduction to this topic discussing both theoretical and practical aspects. It outlines some of the core principles of responsible ML, their importance in applications, and the practical utility of MIO for building ML models that align with these principles. Through examples and mathematical formulations, it illustrates practical strategies and available tools for efficiently solving MIO problems for responsible ML. It concludes with a discussion on current limitations and open research questions, providing suggestions for future work.
|
[
{
"version": "v1",
"created": "Fri, 09 May 2025 07:51:36 GMT"
},
{
"version": "v2",
"created": "Thu, 14 Aug 2025 00:28:10 GMT"
},
{
"version": "v3",
"created": "Sat, 30 Aug 2025 16:26:24 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Justin",
"Nathan",
""
],
[
"Sun",
"Qingshi",
""
],
[
"Gómez",
"Andrés",
""
],
[
"Vayanos",
"Phebe",
""
]
] |
2505.05900
|
Wen-Fan Feng
|
Wen-Fan Feng, Lijing Shao
|
Galactic double neutron stars as dual-line gravitational-wave sources: Prospects with LISA and Cosmic Explorer
|
11 pages, 5 figures, Phys. Rev. D - Accepted 19 July, 2025
|
Phys. Rev. D 112 (2025) 043023
|
10.1103/bqtw-vdws
| null |
gr-qc astro-ph.HE
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Double neutron star (DNS) systems could serve as intriguing dual-line gravitational-wave (GW) sources, emitting both high- and low-frequency GWs, arising respectively from the asymmetric spinning bodies of individual neutron stars (NSs) and the binary orbital inspiral. Detecting such dual-line signals could provide novel perspectives on binary orbital geometry and NS internal physics. We expand upon previously calculated spinning NS waveforms by incorporating higher-order terms of NS structural parameters. A population simulation is performed for spinning NS components in DNS systems potentially detectable by the space-based Laser Interferometer Space Antenna (LISA). Based on 4-year LISA observation of 35 resolvable DNS systems under an optimistic scenario, we estimate that 6 (22) spinning NS components could be detected by the next-generation ground-based GW detector, Cosmic Explorer, under log-uniform (uniform) sampling of NS structural parameters. For these dual-line sources, the median signal-to-noise ratio achievable with Cosmic Explorer is approximately 20--30. Through the dual-line GW detection strategy, the relative measurement accuracy of the NS moment of inertia is estimated to be $\sim 8\%$.
|
[
{
"version": "v1",
"created": "Fri, 09 May 2025 09:07:57 GMT"
},
{
"version": "v2",
"created": "Tue, 22 Jul 2025 14:32:34 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Feng",
"Wen-Fan",
""
],
[
"Shao",
"Lijing",
""
]
] |
2505.06135
|
Adam Freese
|
Adam Freese
|
Mechanical form factors and densities of non-relativistic fermions
|
26 pages, version accepted for publication
|
Phys. Rev. D 112, 034037 (2025)
|
10.1103/k3hd-3lqv
|
JLAB-THY-25-4317
|
hep-ph nucl-th quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The hadron physics community has been actively debating the interpretation of so-called mechanical properties of hadrons. Non-relativistic quantum-mechanical systems like the hydrogen atom have been appealed to in these debates as analogies. Since such appeals are likely to continue, it is important to have Galilei-covariant expressions for matrix elements of the energy-momentum tensor. In this work, I obtain Galilei-covariant breakdowns of such matrix elements into mechanical form factors, with a special focus on spin-half states. I additionally study the spatial densities associated with these form factors, using the pilot wave interpretation to guide their breakdown into contributions from internal structure and from quantum-mechanical effects such as wave packet dispersion. For completeness, I also obtain non-relativistic Breit frame densities.
|
[
{
"version": "v1",
"created": "Fri, 09 May 2025 15:40:49 GMT"
},
{
"version": "v2",
"created": "Fri, 15 Aug 2025 15:36:39 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Freese",
"Adam",
""
]
] |
2505.06247
|
Vakhtang Chulukhadze
|
Vakhtang Chulukhadze, Jack Kramer, Tzu-Hsuan Hsu, Omar Barrera, Ruochen Lu
|
High-Q Millimeter-Wave Acoustic Resonators in Thin-Film Lithium Niobate Using Higher-Order Antisymmetric Modes
|
4 pages with 1 page of reference
|
10.1109/LED.2025.3602762
|
10.1109/LED.2025.3602762
| null |
physics.app-ph
|
http://creativecommons.org/licenses/by/4.0/
|
This letter presents miniature millimeter wave (mmWave, above 30 GHz) acoustic resonators based on a single-layer thin-film lithium niobate (LN) platform. More specifically, we present high performance third-order antisymmetric (A3) mode laterally excited bulk acoustic resonators (XBAR). Compared to prior demonstrations, the proposed platform features a compact footprint due to a smaller lateral wavelength and aperture. We showcase an A3 mode device operating at 39.8 GHz with a high extracted electromechanical coupling (k^2) of 4%, a high 3-dB series resonance quality factor (Q_s) of 97, and a high 3-dB anti-resonance quality factor (Q_p) of 342, leading to a figure of merit (FoM=k^2*Q_p) of 13.7 with a footprint of 32x44 micron^2. To demonstrate frequency scalability, the piezoelectric film thickness is varied while keeping the device layout. As a result, we present a multitude of high-performance devices covering a wide frequency range of 30-50 GHz, validating the proposed XBAR design at mmWave.
|
[
{
"version": "v1",
"created": "Mon, 28 Apr 2025 23:28:56 GMT"
},
{
"version": "v2",
"created": "Thu, 22 May 2025 22:00:59 GMT"
},
{
"version": "v3",
"created": "Tue, 22 Jul 2025 20:32:03 GMT"
},
{
"version": "v4",
"created": "Sat, 30 Aug 2025 17:47:58 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Chulukhadze",
"Vakhtang",
""
],
[
"Kramer",
"Jack",
""
],
[
"Hsu",
"Tzu-Hsuan",
""
],
[
"Barrera",
"Omar",
""
],
[
"Lu",
"Ruochen",
""
]
] |
2505.06331
|
Feilong Jiang Mr.
|
Feilong Jiang, Xiaonan Hou, Jianqiao Ye, Min Xia
|
Mask-PINNs: Mitigating Internal Covariate Shift in Physics-Informed Neural Networks
| null | null | null | null |
cs.LG cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Physics-Informed Neural Networks (PINNs) have emerged as a powerful framework for solving partial differential equations (PDEs) by embedding physical laws directly into the loss function. However, as a fundamental optimization issue, internal covariate shift (ICS) hinders the stable and effective training of PINNs by disrupting feature distributions and limiting model expressiveness. Unlike standard deep learning tasks, conventional remedies for ICS -- such as Batch Normalization and Layer Normalization -- are not directly applicable to PINNs, as they distort the physical consistency required for reliable PDE solutions. To address this issue, we propose Mask-PINNs, a novel architecture that introduces a learnable mask function to regulate feature distributions while preserving the underlying physical constraints of PINNs. We provide a theoretical analysis showing that the mask suppresses the expansion of feature representations through a carefully designed modulation mechanism. Empirically, we validate the method on multiple PDE benchmarks -- including convection, wave propagation, and Helmholtz equations -- across diverse activation functions. Our results show consistent improvements in prediction accuracy, convergence stability, and robustness. Furthermore, we demonstrate that Mask-PINNs enable the effective use of wider networks, overcoming a key limitation in existing PINN frameworks.
|
[
{
"version": "v1",
"created": "Fri, 09 May 2025 15:38:52 GMT"
},
{
"version": "v2",
"created": "Fri, 20 Jun 2025 13:08:04 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 14:38:13 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Jiang",
"Feilong",
""
],
[
"Hou",
"Xiaonan",
""
],
[
"Ye",
"Jianqiao",
""
],
[
"Xia",
"Min",
""
]
] |
2505.06690
|
Jianxin Zhang
|
Jianxin Zhang, Lianzi Jiang, Xinyu Han, Xiangrong Wang
|
A Causality- and Frequency-Aware Deep Learning Framework for Wave Elevation Prediction Behind Floating Breakwaters
| null | null | null | null |
cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Predicting the elevations of nonlinear wave fields behind floating breakwaters (FBs) is crucial for optimizing coastal engineering structures, enhancing safety, and improving design efficiency. Existing deep learning approaches exhibit limited generalization capability under unseen operating conditions. To address this challenge, this study proposes the Exogenous-to-Endogenous Frequency-Aware Network (E2E-FANet), a novel end-to-end neural network designed to model relationships between waves and structures. First, the Dual-Basis Frequency Mapping (DBFM) module leverages orthogonal cosine and sine bases to generate an adaptive time-frequency representation, enabling the model to effectively disentangle the evolving spectral components of wave signals. Second, the Exogenous-to-Endogenous Cross-Attention (E2ECA) module employs cross attention to explicitly model the unidirectional causal influence of floating breakwater motion on wave elevations. Additionally, a Temporal-wise Attention (TA) mechanism is incorporated that adaptively captures complex dependencies in endogenous variables. Extensive experiments, including generalization tests across diverse wave conditions and adaptability tests under varying relative water density (RW) conditions, demonstrate that E2E-FANet achieves superior predictive accuracy and robust generalization compared to mainstream models. This work emphasizes the importance of integrating causality and frequency-aware modeling in deep learning architectures for modeling nonlinear dynamics systems.
|
[
{
"version": "v1",
"created": "Sat, 10 May 2025 16:28:48 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 02:34:10 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Zhang",
"Jianxin",
""
],
[
"Jiang",
"Lianzi",
""
],
[
"Han",
"Xinyu",
""
],
[
"Wang",
"Xiangrong",
""
]
] |
2505.06815
|
Shun-Ya S. Uchida
|
Shun-ya S. Uchida, Suchetha Cooray, Atsushi J. Nishizawa, Tsutomu T. Takeuchi, and Peter Behroozi
|
Insights into the dependence of galaxy properties on the environment with explainable machine learning models
|
accepted to MNRAS
| null | null | null |
astro-ph.GA
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Galaxies reside within dark matter halos, but their properties are influenced not only by their halo properties but also by the surrounding environment. We construct an interpretable neural network framework to characterize the surrounding environment of galaxies and investigate the extent to which their properties are affected by neighboring galaxies in IllustrisTNG300 data ($z=0$). Our models predict galaxy properties (stellar mass and star formation rate) given dark matter subhalo properties of both host subhalo and of surrounding galaxies, which serve as an explainable, flexible galaxy-halo connection model. We find that prediction accuracy peaks when incorporating only the nearest neighboring galaxy for stellar mass prediction, while star formation rate prediction benefits from information from up to the third-nearest neighbor. We determine that environmental influence follows a clear hierarchical pattern, with the nearest neighbor providing the dominant contribution that diminishes rapidly with additional neighbors. We confirm that central and satellite galaxies, as well as different galaxy categories based on mass and star-forming activity, exhibit distinct environmental dependencies. Environmental dependence for low-mass galaxies ($\log(M_*/M_\odot) < 10$) shows 35-50% environmental contribution compared to just 8-30% for massive centrals, while satellite galaxies experience consistently stronger environmental effects than centrals across all populations. Furthermore, we find that the most significant attribute from neighboring subhalos for predicting target galaxy properties is its distance to the nearest neighboring galaxy. These quantitative results offer guidance for constructing more sophisticated empirical and semi-analytic models of galaxy formation that explicitly include environmental dependence as a function of galaxy type and mass.
|
[
{
"version": "v1",
"created": "Sun, 11 May 2025 02:15:42 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 01:18:55 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Uchida",
"Shun-ya S.",
""
],
[
"Cooray",
"Suchetha",
""
],
[
"Nishizawa",
"Atsushi J.",
""
],
[
"Takeuchi",
"Tsutomu T.",
""
],
[
"Behroozi",
"Peter",
""
]
] |
2505.06850
|
Jie Zhao
|
Jie Zhao, Kang Hao Cheong
|
Visual Evolutionary Optimization on Graph-Structured Combinatorial Problems with MLLMs: A Case Study of Influence Maximization
| null | null |
10.1109/TEVC.2025.3598266
| null |
cs.NE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Graph-structured combinatorial problems in complex networks are prevalent in many domains, and are computationally demanding due to their complexity and non-linear nature. Traditional evolutionary algorithms (EAs), while robust, often face obstacles due to content-shallow encoding limitations and lack of structural awareness, necessitating hand-crafted modifications for effective application. In this work, we introduce an original framework, visual evolutionary optimization (VEO), leveraging multimodal large language models (MLLMs) as the backbone evolutionary optimizer in this context. Specifically, we propose a context-aware encoding way, representing the solution of the network as an image. In this manner, we can utilize MLLMs' image processing capabilities to intuitively comprehend network configurations, thus enabling machines to solve these problems in a human-like way. We have developed MLLM-based operators tailored for various evolutionary optimization stages, including initialization, crossover, and mutation. {Furthermore, we propose that graph sparsification can effectively enhance the applicability and scalability of VEO on large-scale networks, owing to the scale-free nature of real-world networks.} We demonstrate the effectiveness of our method using a well-known task in complex networks, influence maximization, and validate it on eight different real-world networks of various structures. The results have confirmed VEO's reliability and enhanced effectiveness compared to traditional evolutionary optimization.
|
[
{
"version": "v1",
"created": "Sun, 11 May 2025 05:23:02 GMT"
},
{
"version": "v2",
"created": "Mon, 11 Aug 2025 06:39:55 GMT"
},
{
"version": "v3",
"created": "Tue, 02 Sep 2025 04:21:57 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Zhao",
"Jie",
""
],
[
"Cheong",
"Kang Hao",
""
]
] |
2505.07058
|
Lakshit Arora
|
Lakshit Arora, Sanjay Surendranath Girija, Shashank Kapoor, Aman Raj, Dipen Pradhan, Ankit Shetgaonkar
|
Explainable Artificial Intelligence Techniques for Software Development Lifecycle: A Phase-specific Survey
|
Accepted to IEEE COMPSAC 2025
|
2025 IEEE 49th Annual Computers, Software, and Applications Conference (COMPSAC)
|
10.1109/COMPSAC65507.2025.00321
| null |
cs.SE cs.AI cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Artificial Intelligence (AI) is rapidly expanding and integrating more into daily life to automate tasks, guide decision making, and enhance efficiency. However, complex AI models, which make decisions without providing clear explanations (known as the "black-box problem"), currently restrict trust and widespread adoption of AI. Explainable Artificial Intelligence (XAI) has emerged to address the black-box problem of making AI systems more interpretable and transparent so stakeholders can trust, verify, and act upon AI-based outcomes. Researchers have developed various techniques to foster XAI in the Software Development Lifecycle. However, there are gaps in applying XAI techniques in the Software Engineering phases. Literature review shows that 68% of XAI in Software Engineering research is focused on maintenance as opposed to 8% on software management and requirements. In this paper, we present a comprehensive survey of the applications of XAI methods such as concept-based explanations, Local Interpretable Model-agnostic Explanations (LIME), SHapley Additive exPlanations (SHAP), rule extraction, attention mechanisms, counterfactual explanations, and example-based explanations to the different phases of the Software Development Life Cycle (SDLC), including requirements elicitation, design and development, testing and deployment, and evolution. To the best of our knowledge, this paper presents the first comprehensive survey of XAI techniques for every phase of the Software Development Life Cycle (SDLC). This survey aims to promote explainable AI in Software Engineering and facilitate the practical application of complex AI models in AI-driven software development.
|
[
{
"version": "v1",
"created": "Sun, 11 May 2025 17:09:57 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Arora",
"Lakshit",
""
],
[
"Girija",
"Sanjay Surendranath",
""
],
[
"Kapoor",
"Shashank",
""
],
[
"Raj",
"Aman",
""
],
[
"Pradhan",
"Dipen",
""
],
[
"Shetgaonkar",
"Ankit",
""
]
] |
2505.07213
|
Wenqiang Zhu
|
Wenqiang Zhu, Xin Wang, Chaoqian Wang, Weijie Xing, Longzhao Liu, Hongwei Zheng, Jingwu Zhao, Shaoting Tang
|
Evolution of cooperation and competition in multilayer networks
|
12 two-column pages, 8 figures
|
Nonlinear Dynamics (2025):1-17
|
10.1007/s11071-025-11693-0
| null |
physics.soc-ph nlin.AO physics.comp-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Cooperation and competition coexist and coevolve in natural and social systems. Cooperation generates resources, which in turn, drive non-cooperative competition to secure individual shares. How this complex interplay between cooperation and competition shapes the evolution of social dilemmas and welfare remains unknown. In this study, we introduce a two-layer evolutionary game model, in which one layer is a cooperative public goods game, and the other is a competitive involution game, with cross-layer feedback linking the two. We find that feedback can either promote or inhibit cooperation, depending on the baseline conditions. For example, moderate resource and synergy factor values can promote social welfare when feedback strength is large. This provides an approach to adjusting the strength and asymmetry of cross-layer feedback to promote cooperation and social welfare. We thus emphasize the importance of managing feedback mechanisms to balance cooperation and competition in complex social systems.
|
[
{
"version": "v1",
"created": "Mon, 12 May 2025 03:46:09 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Zhu",
"Wenqiang",
""
],
[
"Wang",
"Xin",
""
],
[
"Wang",
"Chaoqian",
""
],
[
"Xing",
"Weijie",
""
],
[
"Liu",
"Longzhao",
""
],
[
"Zheng",
"Hongwei",
""
],
[
"Zhao",
"Jingwu",
""
],
[
"Tang",
"Shaoting",
""
]
] |
2505.07371
|
Francesco Scazza
|
Omar Abdel Karim, Alessandro Muzi Falconi, Riccardo Panza, Wenliang Liu, Francesco Scazza
|
Single-atom imaging of ${}^{173}$Yb in optical tweezers loaded by a five-beam magneto-optical trap
|
14 pages, 8 figures
|
Quantum Sci. Technol. 10, 045019 (2025)
|
10.1088/2058-9565/adf7cf
| null |
physics.atom-ph cond-mat.quant-gas quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We report on the trapping and imaging of individual ytterbium atoms in arrays of optical tweezers, loaded from a magneto-optical trap (MOT) formed by only five beams in an orthogonal configuration. In our five-beam MOT, operating on the narrow ${}^1$S${}_0 \rightarrow {}^3$P${}_1$ intercombination transition, gravity balances the radiation pressure of a single upward-directed beam. This approach enables efficient trapping and cooling of the most common ytterbium isotopes (${}^{171}$Yb, ${}^{173}$Yb and ${}^{174}$Yb) to $\lesssim 20\,\mu$K at densities $\sim 10^{11}$ atoms/cm$^3$ within less than one second. This configuration allows for significantly reducing the complexity of the optical setup, potentially benefiting any ytterbium-atom based quantum science platform leveraging single-atom microscopy, from quantum processors to novel optical clocks. We then demonstrate the first single-atom-resolved imaging of the fermionic, large-spin isotope ${}^{173}$Yb ($I=5/2$), employing a two-color imaging scheme that does not rely on magic-wavelength trapping. We achieve a high single-atom detection fidelity of $99.96(1)\%$ and a large survival probability of $98.5(2)\%$, despite large differential light shifts affecting all nuclear spin sublevels of the excited ${}^3$P${}_1$ state involved in the cooling transition. The demonstrated capabilities will play a key role in future quantum simulations and computing applications with ${}^{173}$Yb arrays.
|
[
{
"version": "v1",
"created": "Mon, 12 May 2025 09:14:02 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 08:31:30 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Karim",
"Omar Abdel",
""
],
[
"Falconi",
"Alessandro Muzi",
""
],
[
"Panza",
"Riccardo",
""
],
[
"Liu",
"Wenliang",
""
],
[
"Scazza",
"Francesco",
""
]
] |
2505.07445
|
Alto Osada
|
Alto Osada and Koichiro Miyanishi
|
Integrated Multi-Wavelength Photonic Architectures for Future Scalable Trapped Ion Quantum Devices
|
11 pages, 10 figures including appendix
| null | null | null |
quant-ph physics.app-ph physics.optics
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Recent advances of quantum technologies rely on precise control and integration of quantum objects, and technological breakthrough is anticipated for further scaling up to realize practical applications. Trapped-ion quantum technology is a promising candidates to realize them, while its scalability depends on the development of intra-node scaling up, reproducibility of quantum nodes and photonic interconnection among them. Utilization of integrated photonics instead of free-space optics is a crucial step toward mass production of trapped-ion quantum nodes and manifests itself as useful for laser delivery for various quantum operations and photon detection. However, whole architecture of the scalable photonic circuits for them is left unexplored. In this work, we discuss photonic architectures for trapped-ion quantum devices, in which lasers of multiple wavelengths are delivered to multiple trapping zones within a single chip. We analyze two methods of configuring nanophotonic waveguides and compare them in terms of loss of total laser power. This work opens up a new landscape of photonic architecture for various applications not only to quantum ones such as trapped-ion devices but also to classical photonics, e.g. optoelectronic devices and biochemical sensing.
|
[
{
"version": "v1",
"created": "Mon, 12 May 2025 11:09:31 GMT"
},
{
"version": "v2",
"created": "Wed, 06 Aug 2025 06:38:39 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 04:28:30 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Osada",
"Alto",
""
],
[
"Miyanishi",
"Koichiro",
""
]
] |
2505.07571
|
Marcos Belancon
|
Rafaela Valcarenghi, Brenno Greatti Silva, Robson Ferrari Muniz, Vitor Santaella Zanuto, Anna Paulla Simon, Ricardo Schneider, Raquel Dosciatti Bini, M\'arcio Ant\^onio Fiori, Maxence Vigier, Emmanuel Veron, Mathieu Allix, Marcelo Sandrini, Marcos Paulo Belan\c{c}on
|
Oxyfluoride glasses obtained through incorporation of CaF$_2$ into photovoltaic cover glass melts
|
21 pages, 7 figures
| null | null | null |
cond-mat.mtrl-sci physics.app-ph physics.chem-ph
|
http://creativecommons.org/licenses/by/4.0/
|
The glass industry has limited options to mitigate its environmental footprint, and the demand for cover glass to produce photovoltaic panels is increasing. Currently, the majority of this special type of glass is not being recycled, and in this work, we propose to reuse it as raw material to obtain oxyfluoride glasses. The incorporation of CaF$_2$ and the increasing Na$_2$CO$_3$ content resulted in a melting temperature of about 1200$^o$C, significantly lower than in soda-lime glasses, which adds up to the environmental benefits of reusing end-of-life cover glass. The obtained samples show high transparency and thermal stability, allowing the cover glass to make up to 80\% of its weight. XRF analysis was employed to determine the elemental composition of the samples, while XRD and Raman indicated that by adding CaF$_2$, the glass network was depolymerized. In situ XRD as a function of temperature showed the formation of a few crystalline phases in these oxyfluoride samples, evidencing that it can be explored as a matrix to obtain different glass-ceramics. The combination of the glass properties indicates that this method and the resulting material can contribute to reducing the environmental impact of the glass industry, by creating new glass or glass-ceramic materials that can be obtained at a reduced temperature compared to the soda-lime glass, while cover glass being the primary raw material could reduce the need to extract minerals from nature.
|
[
{
"version": "v1",
"created": "Mon, 12 May 2025 13:52:35 GMT"
},
{
"version": "v2",
"created": "Tue, 20 May 2025 11:27:22 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 18:08:02 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Valcarenghi",
"Rafaela",
""
],
[
"Silva",
"Brenno Greatti",
""
],
[
"Muniz",
"Robson Ferrari",
""
],
[
"Zanuto",
"Vitor Santaella",
""
],
[
"Simon",
"Anna Paulla",
""
],
[
"Schneider",
"Ricardo",
""
],
[
"Bini",
"Raquel Dosciatti",
""
],
[
"Fiori",
"Márcio Antônio",
""
],
[
"Vigier",
"Maxence",
""
],
[
"Veron",
"Emmanuel",
""
],
[
"Allix",
"Mathieu",
""
],
[
"Sandrini",
"Marcelo",
""
],
[
"Belançon",
"Marcos Paulo",
""
]
] |
2505.07776
|
Xinling Li She
|
Xinling Li, Daniele Gammelli, Alex Wallar, Jinhua Zhao, Gioele Zardini
|
Robo-Taxi Fleet Coordination with Accelerated High-Capacity Ridepooling
| null | null | null | null |
math.OC
|
http://creativecommons.org/licenses/by/4.0/
|
Rapid urbanization has led to a surge of customizable mobility demand in urban areas, which makes on-demand services increasingly popular. On-demand services are flexible while reducing the need for private cars, thus mitigating congestion and parking issues in limited urban space. While the coordination of high-capacity ridepooling on-demand service requires effective control to ensure efficiency, the emergence of the paradigm of robo-taxi opens the opportunity for centralized fleet control for an improved service quality. In this work, we propose two acceleration algorithms for the most advanced large-scale high-capacity algorithm proposed in [1]. We prove the improvement in the real-time performance of the algorithm by using real-world on-demand data from Manhattan, NYC.
|
[
{
"version": "v1",
"created": "Mon, 12 May 2025 17:26:32 GMT"
},
{
"version": "v2",
"created": "Sat, 14 Jun 2025 11:01:43 GMT"
},
{
"version": "v3",
"created": "Wed, 27 Aug 2025 15:11:51 GMT"
},
{
"version": "v4",
"created": "Fri, 29 Aug 2025 21:03:25 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Li",
"Xinling",
""
],
[
"Gammelli",
"Daniele",
""
],
[
"Wallar",
"Alex",
""
],
[
"Zhao",
"Jinhua",
""
],
[
"Zardini",
"Gioele",
""
]
] |
2505.08034
|
Lakshit Arora
|
Ankit Shetgaonkar, Dipen Pradhan, Lakshit Arora, Sanjay Surendranath Girija, Shashank Kapoor, Aman Raj
|
Opportunities and Applications of GenAI in Smart Cities: A User-Centric Survey
|
Accepted in IEEE COINS 2025
|
2025 IEEE International Conference on Omni-layer Intelligent Systems (COINS)
|
10.1109/COINS65080.2025.11125750
| null |
cs.OH
|
http://creativecommons.org/licenses/by/4.0/
|
The proliferation of IoT in cities, combined with Digital Twins, creates a rich data foundation for Smart Cities aimed at improving urban life and operations. Generative AI (GenAI) significantly enhances this potential, moving beyond traditional AI analytics and predictions by processing multimodal content and generating novel outputs like text and simulations. Using specialized or foundational models, GenAI's natural language abilities such as Natural Language Understanding (NLU) and Natural Language Generation (NLG) can power tailored applications and unified interfaces, dramatically lowering barriers for users interacting with complex smart city systems. In this paper, we focus on GenAI applications based on conversational interfaces within the context of three critical user archetypes in a Smart City - Citizens, Operators and Planners. We identify and review GenAI models and techniques that have been proposed or deployed for various urban subsystems in the contexts of these user archetypes. We also consider how GenAI can be built on the existing data foundation of official city records, IoT data streams and Urban Digital Twins. We believe this work represents the first comprehensive summarization of GenAI techniques for Smart Cities from the lens of the critical users in a Smart City.
|
[
{
"version": "v1",
"created": "Mon, 12 May 2025 20:04:03 GMT"
},
{
"version": "v2",
"created": "Thu, 14 Aug 2025 23:53:13 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Shetgaonkar",
"Ankit",
""
],
[
"Pradhan",
"Dipen",
""
],
[
"Arora",
"Lakshit",
""
],
[
"Girija",
"Sanjay Surendranath",
""
],
[
"Kapoor",
"Shashank",
""
],
[
"Raj",
"Aman",
""
]
] |
2505.08116
|
Kazumi Okuyama
|
Kazumi Okuyama
|
de Sitter JT gravity from double-scaled SYK
|
18 pages; v2: reference added; v3: to appear in JHEP; v4: published version
| null | null | null |
hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
It is known that the double-scaled SYK model (DSSYK) reduces to JT gravity with a negative cosmological constant by zooming in on the lower edge $E=-E_0$ of the spectrum. We find that the de Sitter JT gravity (i.e. JT gravity with a positive cosmological constant) is reproduced from DSSYK by taking a scaling limit around the upper edge $E=E_0$ of the spectrum. We also argue that the appearance of de Sitter JT gravity is consistent with the behavior of the classical solution of the sine dilaton gravity.
|
[
{
"version": "v1",
"created": "Mon, 12 May 2025 23:12:37 GMT"
},
{
"version": "v2",
"created": "Fri, 16 May 2025 00:28:24 GMT"
},
{
"version": "v3",
"created": "Thu, 31 Jul 2025 23:12:41 GMT"
},
{
"version": "v4",
"created": "Mon, 01 Sep 2025 23:47:27 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Okuyama",
"Kazumi",
""
]
] |
2505.08202
|
Lakshit Arora
|
Aman Raj, Lakshit Arora, Sanjay Surendranath Girija, Shashank Kapoor, Dipen Pradhan, Ankit Shetgaonkar
|
AI and Generative AI Transforming Disaster Management: A Survey of Damage Assessment and Response Techniques
|
Accepted in IEEE Compsac 2025
|
2025 IEEE 49th Annual Computers, Software, and Applications Conference (COMPSAC)
|
10.1109/COMPSAC65507.2025.00251
| null |
cs.CY cs.AI cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Natural disasters, including earthquakes, wildfires and cyclones, bear a huge risk on human lives as well as infrastructure assets. An effective response to disaster depends on the ability to rapidly and efficiently assess the intensity of damage. Artificial Intelligence (AI) and Generative Artificial Intelligence (GenAI) presents a breakthrough solution, capable of combining knowledge from multiple types and sources of data, simulating realistic scenarios of disaster, and identifying emerging trends at a speed previously unimaginable. In this paper, we present a comprehensive review on the prospects of AI and GenAI in damage assessment for various natural disasters, highlighting both its strengths and limitations. We talk about its application to multimodal data such as text, image, video, and audio, and also cover major issues of data privacy, security, and ethical use of the technology during crises. The paper also recognizes the threat of Generative AI misuse, in the form of dissemination of misinformation and for adversarial attacks. Finally, we outline avenues of future research, emphasizing the need for secure, reliable, and ethical Generative AI systems for disaster management in general. We believe that this work represents the first comprehensive survey of Gen-AI techniques being used in the field of Disaster Assessment and Response.
|
[
{
"version": "v1",
"created": "Tue, 13 May 2025 03:33:31 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Raj",
"Aman",
""
],
[
"Arora",
"Lakshit",
""
],
[
"Girija",
"Sanjay Surendranath",
""
],
[
"Kapoor",
"Shashank",
""
],
[
"Pradhan",
"Dipen",
""
],
[
"Shetgaonkar",
"Ankit",
""
]
] |
2505.08545
|
Alexey Dubinsky
|
Alexey Dubinsky
|
Long-Lived Quasinormal Modes and Quasi-Resonances around Non-Minimal Einstein-Yang-Mills Black Holes
| null |
Eur. Phys. J. C 85, 924 (2025)
|
10.1140/epjc/s10052-025-14671-3
| null |
gr-qc
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Using accurate computational methods, we compute the quasinormal frequencies of a massive scalar field propagating near a black hole in the framework of non-minimal Einstein-Yang-Mills theory with a non-zero cosmological constant. We show that increasing the mass of the scalar field significantly decreases the damping rate of the quasinormal modes for both asymptotically flat and de Sitter black holes. However, in the de Sitter case, arbitrarily long-lived modes can exist, whereas in the asymptotically flat case, the damping rate never vanishes completely. In the limit of quasi-resonances, we observe a kind of universal behavior where the frequencies do not depend on the coupling constant. Applying the time-domain integration of perturbation equations we show that even when the effective potential has a negative gap, the scalar field is stable and the perturbations decay in time. In the regime of large mass of the field we obtain the analytic formula for quasinormal modes.
|
[
{
"version": "v1",
"created": "Tue, 13 May 2025 13:17:23 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Dubinsky",
"Alexey",
""
]
] |
2505.08585
|
Jorge Quesada
|
Jorge Quesada and Chen Zhou and Prithwijit Chowdhury and Mohammad Alotaibi and Ahmad Mustafa and Yusufjon Kumakov and Mohit Prabhushankar and Ghassan AlRegib
|
A Large-scale Benchmark on Geological Fault Delineation Models: Domain Shift, Training Dynamics, Generalizability, Evaluation and Inferential Behavior
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Machine learning has taken a critical role in seismic interpretation workflows, especially in fault delineation tasks. However, despite the recent proliferation of pretrained models and synthetic datasets, the field still lacks a systematic understanding of the generalizability limits of these models across seismic data representing diverse geologic, acquisition and processing settings. Distributional shifts between data sources, limitations in fine-tuning strategies and labeled data accessibility, and inconsistent evaluation protocols all remain major roadblocks to deploying reliable models in real-world exploration. In this paper, we present the first large-scale benchmarking study explicitly designed to provide guidelines for domain shift strategies in seismic interpretation. Our benchmark spans over 200 combinations of model architectures, datasets and training strategies, across three datasets (synthetic and real) including FaultSeg3D, CRACKS, and Thebe. We systematically assess pretraining, fine-tuning, and joint training under varying domain shifts. Our analysis shows that common fine-tuning practices can lead to catastrophic forgetting, especially when source and target datasets are disjoint, and that larger models such as Segformer are more robust than smaller architectures. We also find that domain adaptation methods outperform fine-tuning when shifts are large, yet underperform when domains are similar. Finally, we complement segmentation metrics with a novel analysis based on fault characteristic descriptors, revealing how models absorb structural biases from training datasets. Overall, we establish a robust experimental baseline that provides insights into tradeoffs in current fault delineation workflows and highlights directions for building more generalizable and interpretable models.
|
[
{
"version": "v1",
"created": "Tue, 13 May 2025 13:56:43 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 15:20:28 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Quesada",
"Jorge",
""
],
[
"Zhou",
"Chen",
""
],
[
"Chowdhury",
"Prithwijit",
""
],
[
"Alotaibi",
"Mohammad",
""
],
[
"Mustafa",
"Ahmad",
""
],
[
"Kumakov",
"Yusufjon",
""
],
[
"Prabhushankar",
"Mohit",
""
],
[
"AlRegib",
"Ghassan",
""
]
] |
2505.08615
|
Jan Ditzen
|
Jan Ditzen and Ovidijus Stauskas
|
On Selection of Cross-Section Averages in Non-stationary Environments
| null | null | null | null |
econ.EM
|
http://creativecommons.org/licenses/by/4.0/
|
Information criteria (IC) have been widely used in factor models to estimate an unknown number of latent factors. It has recently been shown that IC perform well in Common Correlated Effects (CCE) and related setups in selecting a set of cross-section averages (CAs) sufficient for the factor space under stationary factors. As CAs can proxy non-stationary factors, it is tempting to claim such generality of IC, too. We show formally and in simulations that IC have a severe underselection issue even under very mild forms of factor non-stationarity, which goes against the sentiment in the literature.
|
[
{
"version": "v1",
"created": "Tue, 13 May 2025 14:32:37 GMT"
},
{
"version": "v2",
"created": "Tue, 20 May 2025 11:44:11 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 11:02:47 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Ditzen",
"Jan",
""
],
[
"Stauskas",
"Ovidijus",
""
]
] |
2505.08667
|
Anna Zakharova
|
Anna Zakharova
|
$\mathcal{P}$, $\mathcal{T}$-violating axion-mediated interactions in RaOH molecule
|
15 pages, 4 figures
| null | null | null |
hep-ph physics.atom-ph quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
If axion simultaneously has the scalar couplings to the nucleons and pseudo-scalar couplings to the electrons, it may mediate a $\mathcal{P}$, $\mathcal{T}$-violating interaction between the electronic shell and nuclei in the molecules. The polyatomic molecule RaOH, which is considered as a promising platform for the $\mathcal{P}$, $\mathcal{T}$ violation searches, is studied for its sensitivity to such interactions. Due to the long-range nature (on molecular scales) of the axion-mediated interaction, it is important whether the enhancement parameter would be sensitive to the vibration of the molecule. Our results imply that the impact of the vibrations on the axion-mediated electron-nucleon interaction in the molecule is similar to the impact on the short-range electron-nucleon scalar-pseudoscalar interaction studied earlier.
|
[
{
"version": "v1",
"created": "Tue, 13 May 2025 15:29:39 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 15:25:13 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Zakharova",
"Anna",
""
]
] |
2505.09148
|
Barnali Das
|
Barnali Das, Laura N. Driessen, Matt E. Shultz, Joshua Pritchard, Kovi Rose, Yuanming Wang, Yu Wing Joshua Lee, Gregory Sivakoff, Andrew Zic and Tara Murphy
|
VAST-MeMeS: Characterising non-thermal radio emission from magnetic massive stars using the Australian SKA Pathfinder
|
Accepted for publication in PASA
| null | null | null |
astro-ph.SR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Magnetic massive stars are stars of spectral types O, B and A that harbour $\sim$ kG strength (mostly dipolar) surface magnetic fields. Their non-thermal radio emission has been demonstrated to be an important magnetospheric probe, provided the emission is fully characterised. A necessary step for that is to build a statistically significant sample of radio-bright magnetic massive stars. In this paper, we present the `VAST project to study Magnetic Massive Stars' or VAST-MeMeS that aims to achieve that by taking advantage of survey data acquired with the Australian SKA Pathfinder telescope. VAST-MeMeS is defined under the `VAriable and Slow Transient' (VAST) survey, although it also uses data from other ASKAP surveys. We found radio detections from 48 magnetic massive stars, out of which, 14 do not have any prior radio detections. We also identified 9 `Main-sequence Radio Pulse Emitter' candidates based on variability and circular polarisation of flux densities. The expanded sample suggests a slightly lower efficiency in the radio production than that reported in earlier work. In addition to significantly expanding the sample of radio-bright magnetic massive stars, the addition of flux density measurements at $\lesssim 1$ GHz revealed that the spectra of incoherent radio emission can extend to much lower frequencies than that assumed in the past. In the future, radio observations spanning wide frequency and rotational phase ranges should be conducted so as to reduce the uncertainties in the incoherent radio luminosities. The results from these campaigns, supplemented with precise estimations of stellar parameters, will allow us to fully understand particle acceleration and non-thermal radio production in large-scale stellar magnetospheres.
|
[
{
"version": "v1",
"created": "Wed, 14 May 2025 05:10:03 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 03:27:49 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Das",
"Barnali",
""
],
[
"Driessen",
"Laura N.",
""
],
[
"Shultz",
"Matt E.",
""
],
[
"Pritchard",
"Joshua",
""
],
[
"Rose",
"Kovi",
""
],
[
"Wang",
"Yuanming",
""
],
[
"Lee",
"Yu Wing Joshua",
""
],
[
"Sivakoff",
"Gregory",
""
],
[
"Zic",
"Andrew",
""
],
[
"Murphy",
"Tara",
""
]
] |
2505.09219
|
Yong Tang
|
Mu-Chun Chen and Yong Tang
|
Probing Self-Interacting Dark Matter via Gravitational-Wave Background from Eccentric Supermassive Black Hole Mergers
|
16 pages, 7 figures; Accepted on Phys. Rev. D
|
Phys.Rev.D 112 (2025) 4, 043033
|
10.1103/lzqc-xr4c
| null |
astro-ph.GA astro-ph.CO gr-qc hep-ph
|
http://creativecommons.org/licenses/by/4.0/
|
The nature of dark matter is still mysterious despite various astronomical evidence. As a possible candidate, self-interacting dark matter (SIDM) can potentially resolve some issues appearing in cold dark matter paradigm. Here we investigate how SIDM around supermassive black holes (SMBH) in galaxy centers may form a density spike and imprint in the spectrum shape of stochastic gravitational-wave background from SMBH binaries (SMBHBs). Employing a refined dynamical friction formula and consistently evolving the orbital dynamics, we demonstrate that current pulsar timing arrays (PTAs) data is sensitive to the cross section of SIDM with $\sigma(v)/m_\chi\lesssim0.66\,\mathrm{cm}^2/\mathrm{g}$, comparable to other astrophysical probes. We also highlight the importance of including the eccentricity of SMBHBs in the parameter inference, which would affect the results significantly. Our findings reveal the promising potential of PTAs observations in probing the nature of dark matter.
|
[
{
"version": "v1",
"created": "Wed, 14 May 2025 08:40:19 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 11:42:34 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Chen",
"Mu-Chun",
""
],
[
"Tang",
"Yong",
""
]
] |
2505.09260
|
Pratibha Raghupati Hegde Ms
|
Pratibha Raghupati Hegde, Paolo Marcandelli, Yuanchun He, Luca Pennati, Jeremy J. Williams, Ivy Peng, Stefano Markidis
|
A Hybrid Quantum-Classical Particle-in-Cell Method for Plasma Simulations
|
10 pages, 9 figures
| null |
10.1016/j.future.2025.108087
| null |
quant-ph cs.ET
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present a hybrid quantum-classical electrostatic Particle-in-Cell (PIC) method, where the electrostatic field Poisson solver is implemented on a quantum computer simulator using a hybrid classical-quantum Neural Network (HNN) using data-driven and physics-informed learning approaches. The HNN is trained on classical PIC simulation results and executed via a PennyLane quantum simulator. The remaining computational steps, including particle motion and field interpolation, are performed on a classical system. To evaluate the accuracy and computational cost of this hybrid approach, we test the hybrid quantum-classical electrostatic PIC against the two-stream instability, a standard benchmark in plasma physics. Our results show that the quantum Poisson solver achieves comparable accuracy to classical methods. It also provides insights into the feasibility of using quantum computing and HNNs for plasma simulations. We also discuss the computational overhead associated with current quantum computer simulators, showing the challenges and potential advantages of hybrid quantum-classical numerical methods.
|
[
{
"version": "v1",
"created": "Wed, 14 May 2025 10:19:33 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 13:04:46 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Hegde",
"Pratibha Raghupati",
""
],
[
"Marcandelli",
"Paolo",
""
],
[
"He",
"Yuanchun",
""
],
[
"Pennati",
"Luca",
""
],
[
"Williams",
"Jeremy J.",
""
],
[
"Peng",
"Ivy",
""
],
[
"Markidis",
"Stefano",
""
]
] |
2505.09288
|
Simon Krekels
|
Simon Krekels, Vukan Levajac, Kristof Moors, George Simion, Bart Sor\'ee
|
Current Conservation in the Self-Consistent Josephson Junction
|
Updated version
| null | null | null |
cond-mat.supr-con cond-mat.mes-hall
|
http://creativecommons.org/licenses/by/4.0/
|
Conventional treatments of Josephson junctions (JJs) are typically not current-conserving. In the mean-field BCS theory, current conservation is only guaranteed if the superconducting order parameter is treated self-consistently. We show that this requirement has significant consequences for the current-phase relation (CPR) in certain regimes, where the current density in the superconducting leads is non-negligible. To this end, we introduce a numerical method for the self-consistent treatment of the BdG equations with current conservation for quasi-1D superconductor-normal (metal)-superconductor (SNS) JJs. Our model incorporates a phase gradient of the order parameter in the leads, which is set to match the Josephson current through the weak link. We compare our method to standard, non-current-conserving approaches by calculating the CPR for SNS JJs while varying lengths and gate voltages controlling the normal metal. We show that current conservation has significant implications for the Josephson harmonics and can weaken or even reverse forward skewness of the CPR.
|
[
{
"version": "v1",
"created": "Wed, 14 May 2025 11:13:35 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 14:02:42 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Krekels",
"Simon",
""
],
[
"Levajac",
"Vukan",
""
],
[
"Moors",
"Kristof",
""
],
[
"Simion",
"George",
""
],
[
"Sorée",
"Bart",
""
]
] |
2505.09365
|
Hannah Theresa R\"udisser
|
H. T. R\"udisser, G. Nguyen, J. Le Lou\"edec, E. E. Davies, C. M\"ostl
|
ARCANE -- Early Detection of Interplanetary Coronal Mass Ejections
|
29 pages, 10 figures, 1 table, submitted to AGU Space Weather on 14th May 2025, revised 29th August 2025
| null | null | null |
physics.space-ph astro-ph.IM astro-ph.SR cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Interplanetary coronal mass ejections (ICMEs) are major drivers of space weather disturbances, posing risks to both technological infrastructure and human activities. Automatic detection of ICMEs in solar wind in situ data is essential for early warning systems. While several methods have been proposed to identify these structures in time series data, robust real-time detection remains a significant challenge. In this work, we present ARCANE - the first framework explicitly designed for early ICME detection in streaming solar wind data under realistic operational constraints, enabling event identification without requiring observation of the full structure. Our approach evaluates the strengths and limitations of detection models by comparing a machine learning-based method to a threshold-based baseline. The ResUNet++ model, previously validated on science data, significantly outperforms the baseline, particularly in detecting high-impact events, while retaining solid performance on lower-impact cases. Notably, we find that using real-time solar wind (RTSW) data instead of high-resolution science data leads to only minimal performance degradation. Despite the challenges of operational settings, our detection pipeline achieves an F1-Score of 0.37, with an average detection delay of 24.5% of the event's duration while processing only a minimal portion of the event data. As more data becomes available, the performance increases significantly. These results mark a substantial step forward in automated space weather monitoring and lay the groundwork for enhanced real-time forecasting capabilities.
|
[
{
"version": "v1",
"created": "Wed, 14 May 2025 13:17:45 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 10:56:28 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Rüdisser",
"H. T.",
""
],
[
"Nguyen",
"G.",
""
],
[
"Louëdec",
"J. Le",
""
],
[
"Davies",
"E. E.",
""
],
[
"Möstl",
"C.",
""
]
] |
2505.09554
|
Ioakeim Ampatzoglou
|
Ioakeim Ampatzoglou, Tristan L\'eger
|
Convolution estimates for the Boltzmann gain operator with hard spheres
|
36 pages. The convolution estimates on the gain operator have been improved. Global existence will be addressed in a separate paper
| null | null | null |
math.AP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We prove new moment-preserving polynomially weighted convolution estimates for the gain operator of the Boltzmann equation with hard potentials, including the critical case of hard-spheres. Our approach relies crucially on a novel cancellation mechanism dealing with the pathological case of energy-absorbing collisions (that is, collisions that accumulate energy to only one of the outgoing particles). These collisions distinguish hard potentials from Maxwell molecules. Our method quantifies the heuristic that, while energy-absorbing collisions occur with non-trivial probability, they are statistically rare, and therefore do not affect the overall averaging behavior of the gain operator. At the technical level, our proof relies solely on tools from kinetic theory, such as geometric identities and angular averaging.
|
[
{
"version": "v1",
"created": "Wed, 14 May 2025 16:53:10 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 15:00:02 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Ampatzoglou",
"Ioakeim",
""
],
[
"Léger",
"Tristan",
""
]
] |
2505.09698
|
Daniel Seita
|
Enyu Zhao, Vedant Raval, Hejia Zhang, Jiageng Mao, Zeyu Shangguan, Stefanos Nikolaidis, Yue Wang, Daniel Seita
|
ManipBench: Benchmarking Vision-Language Models for Low-Level Robot Manipulation
|
Conference on Robot Learning (CoRL) 2025. 50 pages and 30 figures. v2 is the camera-ready and includes a few more new experiments compared to v1
| null | null | null |
cs.RO cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Vision-Language Models (VLMs) have revolutionized artificial intelligence and robotics due to their commonsense reasoning capabilities. In robotic manipulation, VLMs are used primarily as high-level planners, but recent work has also studied their lower-level reasoning ability, which refers to making decisions about precise robot movements. However, the community currently lacks a clear and common benchmark that can evaluate how well VLMs can aid low-level reasoning in robotics. Consequently, we propose a novel benchmark, ManipBench, to evaluate the low-level robot manipulation reasoning capabilities of VLMs across various dimensions, including how well they understand object-object interactions and deformable object manipulation. We extensively test 33 representative VLMs across 10 model families on our benchmark, including variants to test different model sizes. Our evaluation shows that the performance of VLMs significantly varies across tasks, and there is a strong correlation between this performance and trends in our real-world manipulation tasks. It also shows that there remains a significant gap between these models and human-level understanding. See our website at: https://manipbench.github.io.
|
[
{
"version": "v1",
"created": "Wed, 14 May 2025 18:01:00 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 18:59:30 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Zhao",
"Enyu",
""
],
[
"Raval",
"Vedant",
""
],
[
"Zhang",
"Hejia",
""
],
[
"Mao",
"Jiageng",
""
],
[
"Shangguan",
"Zeyu",
""
],
[
"Nikolaidis",
"Stefanos",
""
],
[
"Wang",
"Yue",
""
],
[
"Seita",
"Daniel",
""
]
] |
2505.09958
|
Huaiyu Wang
|
Huaiyu Wang, Vladimir Stoica, Cheng Dai, Marek Pa\'sciak, Sujit Das, Tiannan Yang, Mauro A. P. Gon\c{c}alves, Jiri Kulda, Margaret R. McCarter, Anudeep Mangu, Yue Cao, Hari Padma, Utkarsh Saha, Diling Zhu, Takahiro Sato, Sanghoon Song, Mathias Hoffmann, Patrick Kramer, Silke Nelson, Yanwen Sun, Quynh Nguyen, Zhan Zhang, Ramamoorthy Ramesh, Lane Martin, Aaron M. Lindenberg, Long-Qing Chen, John W. Freeland, Jirka Hlinka, Venkatraman Gopalan, Haidan Wen
|
Terahertz-field activation of polar skyrons
| null | null | null | null |
cond-mat.mtrl-sci physics.app-ph
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Unraveling collective modes arising from coupled degrees of freedom is crucial for understanding complex interactions in solids and developing new functionalities. Unique collective behaviors emerge when two degrees of freedom, ordered on distinct length scales, interact. Polar skyrmions, three-dimensional electric polarization textures in ferroelectric superlattices, disrupt the lattice continuity at the nanometer scale with nontrivial topology, leading to previously unexplored collective modes. Here, using terahertz-field excitation and femtosecond x-ray diffraction, we discovered subterahertz collective modes, dubbed 'skyrons', which appear as swirling patterns of atomic displacements functioning as atomic-scale gearsets. Momentum-resolved time-domain measurements of diffuse scattering revealed an avoided crossing in the dispersion relation of skyrons. We further demonstrated that the amplitude and dispersion of skyrons can be controlled by sample temperature and electric-field bias. Atomistic simulations and dynamical phase-field modeling provided microscopic insights into the three-dimensional crystallographic and polarization dynamics. The discovery of skyrons and their coupling with terahertz fields opens avenues for ultrafast control of topological polar structures.
|
[
{
"version": "v1",
"created": "Thu, 15 May 2025 04:41:06 GMT"
},
{
"version": "v2",
"created": "Thu, 19 Jun 2025 22:01:02 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 18:06:12 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Wang",
"Huaiyu",
""
],
[
"Stoica",
"Vladimir",
""
],
[
"Dai",
"Cheng",
""
],
[
"Paściak",
"Marek",
""
],
[
"Das",
"Sujit",
""
],
[
"Yang",
"Tiannan",
""
],
[
"Gonçalves",
"Mauro A. P.",
""
],
[
"Kulda",
"Jiri",
""
],
[
"McCarter",
"Margaret R.",
""
],
[
"Mangu",
"Anudeep",
""
],
[
"Cao",
"Yue",
""
],
[
"Padma",
"Hari",
""
],
[
"Saha",
"Utkarsh",
""
],
[
"Zhu",
"Diling",
""
],
[
"Sato",
"Takahiro",
""
],
[
"Song",
"Sanghoon",
""
],
[
"Hoffmann",
"Mathias",
""
],
[
"Kramer",
"Patrick",
""
],
[
"Nelson",
"Silke",
""
],
[
"Sun",
"Yanwen",
""
],
[
"Nguyen",
"Quynh",
""
],
[
"Zhang",
"Zhan",
""
],
[
"Ramesh",
"Ramamoorthy",
""
],
[
"Martin",
"Lane",
""
],
[
"Lindenberg",
"Aaron M.",
""
],
[
"Chen",
"Long-Qing",
""
],
[
"Freeland",
"John W.",
""
],
[
"Hlinka",
"Jirka",
""
],
[
"Gopalan",
"Venkatraman",
""
],
[
"Wen",
"Haidan",
""
]
] |
2505.10079
|
Deung-Jang Choi
|
Deung-Jang Choi, Soo-hyon Phark, Andreas J. Heinrich, and Nicol\'as Lorente
|
Electron spin resonance with scanning tunneling microscopy: a tool for an on-surface quantum platform of identical qubits
| null |
Nanoscale Advances, 7, 4551-4558 (2025)
|
10.1039/D5NA00316D
| null |
cond-mat.mes-hall quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Integration of electron spin resonance (ESR) in a scanning tunneling microscope (STM) has enabled an all-electrical control of atomic and molecular spins on solid surfaces with atomic-scale precision and energy resolution beyond thermal limitations. Further, coherent manipulation and detection of individual spins in an ESR-STM establishes a powerful quantum platform, allowing for the implementation of fundamental quantum logic operations to on-surface identical qubits. In this review, we introduce recent advances of ESR-STM, focusing on its application to atomic-scale qubits and extension to molecular qubit systems. We discuss the principles underlying ESR-STM, followed by single-spin addressability, coherent control via Rabi oscillations, and quantum state readout through frequency-resolved detection. We further demonstrate multi-qubit control architectures enabled by atom manipulation and local magnetic field engineering, culminating in the realization of multi-qubit logic gates such as the Controlled-NOT and Toffoli gates. These implementations highlight the specialty of ESR-STM towards atomic-scale quantum circuits. Indeed, ESR-STM can be an excellent tool to perform and evaluate quantum operations in molecular qubits. The results reviewed in this collection establish ESR-STM as a versatile tool for advancing quantum coherent science at the atomic and molecular level in solid-state environments.
|
[
{
"version": "v1",
"created": "Thu, 15 May 2025 08:34:53 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Choi",
"Deung-Jang",
""
],
[
"Phark",
"Soo-hyon",
""
],
[
"Heinrich",
"Andreas J.",
""
],
[
"Lorente",
"Nicolás",
""
]
] |
2505.10292
|
Daniel Oliveira
|
Daniel A. P. Oliveira, David Martins de Matos
|
StoryReasoning Dataset: Using Chain-of-Thought for Scene Understanding and Grounded Story Generation
|
31 pages, 14 figures
| null | null | null |
cs.CV cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Visual storytelling systems struggle to maintain character identity across frames and link actions to appropriate subjects, frequently leading to referential hallucinations. These issues can be addressed through grounding of characters, objects, and other entities on the visual elements. We propose StoryReasoning, a dataset containing 4,178 stories derived from 52,016 movie images, with both structured scene analyses and grounded stories. Each story maintains character and object consistency across frames while explicitly modeling multi-frame relationships through structured tabular representations. Our approach features cross-frame object re-identification using visual similarity and face recognition, chain-of-thought reasoning for explicit narrative modeling, and a grounding scheme that links textual elements to visual entities across multiple frames. We establish baseline performance by fine-tuning Qwen2.5-VL 7B, creating Qwen Storyteller, which performs end-to-end object detection, re-identification, and landmark detection while maintaining consistent object references throughout the story. Evaluation demonstrates a reduction from 4.06 to 3.56 (-12.3%) hallucinations on average per story and an improvement in creativity from 2.58 to 3.38 (+31.0%) when compared to a non-fine-tuned model.
|
[
{
"version": "v1",
"created": "Thu, 15 May 2025 13:42:14 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 14:31:15 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Oliveira",
"Daniel A. P.",
""
],
[
"de Matos",
"David Martins",
""
]
] |
2505.10618
|
Petr Jakub\v{c}\'ik
|
Xuan Chen, Petr Jakub\v{c}\'ik, Matteo Marcoli, Giovanni Stagnitto
|
Jet production at electron-positron colliders at next-to-next-to-next-to-leading order in QCD
|
10 pages, 3 figures; v2 matches published version
|
Phys.Lett.B 869 (2025) 139804
|
10.1016/j.physletb.2025.139804
|
IPPP/25/27, ZU-TH 36/25
|
hep-ph
|
http://creativecommons.org/licenses/by/4.0/
|
We present the first application of antenna subtraction at next-to-next-to-next-to-leading order (N$^3$LO) in QCD by computing fully differential predictions for two-jet production at electron-positron colliders. We illustrate the structure of the infrared counterterms and provide results for the N$^3$LO correction to the two-jet production rate and to the leading-jet energy. Our work constitutes the first direct calculation of jet production at electron-positron colliders at N$^3$LO and represents the first step in tackling arbitrary processes with jets at this perturbative order.
|
[
{
"version": "v1",
"created": "Thu, 15 May 2025 18:00:02 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 09:44:40 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Chen",
"Xuan",
""
],
[
"Jakubčík",
"Petr",
""
],
[
"Marcoli",
"Matteo",
""
],
[
"Stagnitto",
"Giovanni",
""
]
] |
2505.10974
|
Taras Hutak
|
Taras Hutak
|
Thermodynamics of the $S=1/2$ maple-leaf Heisenberg antiferromagnet
|
9 pages, 5 figures
| null | null | null |
cond-mat.str-el
|
http://creativecommons.org/licenses/by/4.0/
|
The Heisenberg antiferromagnet on the maple-leaf lattice has recently gathered a great deal of attention. Competition between three nonequivalent bond interactions results in various ground-state quantum phases, the exact dimer-product singlet ground state being among them. The thermodynamic properties of this model are much less understood. We used high-temperature expansion up to the $18$th order to study the thermodynamics of the $S=1/2$ Heisenberg model on the uniform maple-leaf lattice with the ground state exhibiting a six-sublattice $120^{\circ}$ long-range magnetic order. Pad\'{e} approximants allow us to get reliable results up to the temperatures of about $T\approx 0.4$. To study thermodynamics for arbitrary temperatures, we made the interpolation using the entropy method. Based on the analysis of close Pad\'{e} approximants, we find ground-state energy $e_{0}=-0.53064\ldots -0.53023$ in good agreement with numerical results. The specific heat $c(T)$ has a typical maximum at rather low temperatures $T\approx0.379$ and the uniform susceptibility $\chi(T)$ at $T\approx0.49$. We also estimate the value of $\chi(T)$ at zero temperature $\chi_{0}\approx0.05\ldots0.06$. The ground-state order manifests itself in the divergence of the so-called generalized Wilson ratio.
|
[
{
"version": "v1",
"created": "Fri, 16 May 2025 08:21:21 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 17:37:44 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Hutak",
"Taras",
""
]
] |
2505.11277
|
Yaorui Shi
|
Yaorui Shi, Sihang Li, Chang Wu, Zhiyuan Liu, Junfeng Fang, Hengxing Cai, An Zhang, Xiang Wang
|
Search and Refine During Think: Autonomous Retrieval-Augmented Reasoning of LLMs
| null | null | null | null |
cs.CL cs.AI
|
http://creativecommons.org/licenses/by-sa/4.0/
|
Large language models have demonstrated impressive reasoning capabilities but are inherently limited by their knowledge reservoir. Retrieval-augmented reasoning mitigates this limitation by allowing LLMs to query external resources, but existing methods often retrieve irrelevant or noisy information, hindering accurate reasoning. In this paper, we propose AutoRefine, a reinforcement learning post-training framework that adopts a new ``search-and-refine-during-think'' paradigm. AutoRefine introduces explicit knowledge refinement steps between successive search calls, enabling the model to iteratively filter, distill, and organize evidence before generating an answer. Furthermore, we incorporate tailored retrieval-specific rewards alongside answer correctness rewards using group relative policy optimization. Experiments on single-hop and multi-hop QA benchmarks demonstrate that AutoRefine significantly outperforms existing approaches, particularly in complex, multi-hop reasoning scenarios. Detailed analysis shows that AutoRefine issues frequent, higher-quality searches and synthesizes evidence effectively.
|
[
{
"version": "v1",
"created": "Fri, 16 May 2025 14:11:29 GMT"
},
{
"version": "v2",
"created": "Wed, 28 May 2025 02:19:51 GMT"
},
{
"version": "v3",
"created": "Thu, 26 Jun 2025 06:52:37 GMT"
},
{
"version": "v4",
"created": "Sun, 31 Aug 2025 11:44:41 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Shi",
"Yaorui",
""
],
[
"Li",
"Sihang",
""
],
[
"Wu",
"Chang",
""
],
[
"Liu",
"Zhiyuan",
""
],
[
"Fang",
"Junfeng",
""
],
[
"Cai",
"Hengxing",
""
],
[
"Zhang",
"An",
""
],
[
"Wang",
"Xiang",
""
]
] |
2505.11310
|
ATLAS Collaboration
|
ATLAS Collaboration
|
Measurements of $W^+W^-$ production cross-sections in $pp$ collisions at $\sqrt{s}=13$ TeV with the ATLAS detector
|
62 pages in total, author list starting page 45, 15 figures, 7 tables, submitted to JHEP. All figures including auxiliary figures are available at https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/PAPERS/STDM-2020-16/
|
JHEP 08 (2025) 142
|
10.1007/JHEP08(2025)142
|
CERN-EP-2025-091
|
hep-ex
|
http://creativecommons.org/licenses/by/4.0/
|
Measurements of $W^+W^-\rightarrow e^\pm \nu \mu^\mp \nu$ production cross-sections are presented, providing a test of the predictions of perturbative quantum chromodynamics and the electroweak theory. The measurements are based on data from $pp$ collisions at $\sqrt{s}=13$ TeV recorded by the ATLAS detector at the Large Hadron Collider in 2015-2018, corresponding to an integrated luminosity of 140 fb$^{-1}$. The number of events due to top-quark pair production, the largest background, is reduced by rejecting events containing jets with $b$-hadron decays. An improved methodology for estimating the remaining top-quark background enables a precise measurement of $W^+W^-$ cross-sections with no additional requirements on jets. The fiducial $W^+W^-$ cross-section is determined in a maximum-likelihood fit with an uncertainty of 3.1%. The measurement is extrapolated to the full phase space, resulting in a total $W^+W^-$ cross-section of $127\pm4$ pb. Differential cross-sections are measured as a function of twelve observables that comprehensively describe the kinematics of $W^+W^-$ events. The measurements are compared with state-of-the-art theory calculations and excellent agreement with predictions is observed. A charge asymmetry in the lepton rapidity is observed as a function of the dilepton invariant mass, in agreement with the Standard Model expectation. A CP-odd observable is measured to be consistent with no CP violation. Limits on Standard Model effective field theory Wilson coefficients in the Warsaw basis are obtained from the differential cross-sections.
|
[
{
"version": "v1",
"created": "Fri, 16 May 2025 14:35:52 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"ATLAS Collaboration",
"",
""
]
] |
2505.11318
|
Donald Loveland
|
Donald Loveland, Mingxuan Ju, Tong Zhao, Neil Shah, Danai Koutra
|
On the Role of Weight Decay in Collaborative Filtering: A Popularity Perspective
|
Accepted at SIGKDD 2025
| null | null | null |
cs.IR cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Collaborative filtering (CF) enables large-scale recommendation systems by encoding information from historical user-item interactions into dense ID-embedding tables. However, as embedding tables grow, closed-form solutions become impractical, often necessitating the use of mini-batch gradient descent for training. Despite extensive work on designing loss functions to train CF models, we argue that one core component of these pipelines is heavily overlooked: weight decay. Attaining high-performing models typically requires careful tuning of weight decay, regardless of loss, yet its necessity is not well understood. In this work, we question why weight decay is crucial in CF pipelines and how it impacts training. Through theoretical and empirical analysis, we surprisingly uncover that weight decay's primary function is to encode popularity information into the magnitudes of the embedding vectors. Moreover, we find that tuning weight decay acts as a coarse, non-linear knob to influence preference towards popular or unpopular items. Based on these findings, we propose PRISM (Popularity-awaRe Initialization Strategy for embedding Magnitudes), a straightforward yet effective solution to simplify the training of high-performing CF models. PRISM pre-encodes the popularity information typically learned through weight decay, eliminating its necessity. Our experiments show that PRISM improves performance by up to 4.77% and reduces training times by 38.48%, compared to state-of-the-art training strategies. Additionally, we parameterize PRISM to modulate the initialization strength, offering a cost-effective and meaningful strategy to mitigate popularity bias.
|
[
{
"version": "v1",
"created": "Fri, 16 May 2025 14:41:57 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 21:01:57 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Loveland",
"Donald",
""
],
[
"Ju",
"Mingxuan",
""
],
[
"Zhao",
"Tong",
""
],
[
"Shah",
"Neil",
""
],
[
"Koutra",
"Danai",
""
]
] |
2505.11423
|
Xiaomin Li
|
Xiaomin Li, Zhou Yu, Zhiwei Zhang, Xupeng Chen, Ziji Zhang, Yingying Zhuang, Narayanan Sadagopan, Anurag Beniwal
|
When Thinking Fails: The Pitfalls of Reasoning for Instruction-Following in LLMs
| null | null | null | null |
cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Reasoning-enhanced large language models (RLLMs), whether explicitly trained for reasoning or prompted via chain-of-thought (CoT), have achieved state-of-the-art performance on many complex reasoning tasks. However, we uncover a surprising and previously overlooked phenomenon: explicit CoT reasoning can significantly degrade instruction-following accuracy. Evaluating 15 models on two benchmarks: IFEval (with simple, rule-verifiable constraints) and ComplexBench (with complex, compositional constraints), we consistently observe performance drops when CoT prompting is applied. Through large-scale case studies and an attention-based analysis, we identify common patterns where reasoning either helps (e.g., with formatting or lexical precision) or hurts (e.g., by neglecting simple constraints or introducing unnecessary content). We propose a metric, constraint attention, to quantify model focus during generation and show that CoT reasoning often diverts attention away from instruction-relevant tokens. To mitigate these effects, we introduce and evaluate four strategies: in-context learning, self-reflection, self-selective reasoning, and classifier-selective reasoning. Our results demonstrate that selective reasoning strategies, particularly classifier-selective reasoning, can substantially recover lost performance. To our knowledge, this is the first work to systematically expose reasoning-induced failures in instruction-following and offer practical mitigation strategies.
|
[
{
"version": "v1",
"created": "Fri, 16 May 2025 16:36:00 GMT"
},
{
"version": "v2",
"created": "Tue, 20 May 2025 05:31:43 GMT"
},
{
"version": "v3",
"created": "Tue, 02 Sep 2025 02:55:17 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Li",
"Xiaomin",
""
],
[
"Yu",
"Zhou",
""
],
[
"Zhang",
"Zhiwei",
""
],
[
"Chen",
"Xupeng",
""
],
[
"Zhang",
"Ziji",
""
],
[
"Zhuang",
"Yingying",
""
],
[
"Sadagopan",
"Narayanan",
""
],
[
"Beniwal",
"Anurag",
""
]
] |
2505.11548
|
Zhiyuan Chang
|
Zhiyuan Chang, Mingyang Li, Xiaojun Jia, Junjie Wang, Yuekai Huang, Ziyou Jiang, Yang Liu, Qing Wang
|
One Shot Dominance: Knowledge Poisoning Attack on Retrieval-Augmented Generation Systems
|
15pages, 4 figures
| null | null | null |
cs.CR cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Large Language Models (LLMs) enhanced with Retrieval-Augmented Generation (RAG) have shown improved performance in generating accurate responses. However, the dependence on external knowledge bases introduces potential security vulnerabilities, particularly when these knowledge bases are publicly accessible and modifiable. While previous studies have exposed knowledge poisoning risks in RAG systems, existing attack methods suffer from critical limitations: they either require injecting multiple poisoned documents (resulting in poor stealthiness) or can only function effectively on simplistic queries (limiting real-world applicability). This paper reveals a more realistic knowledge poisoning attack against RAG systems that achieves successful attacks by poisoning only a single document while remaining effective for complex multi-hop questions involving complex relationships between multiple elements. Our proposed AuthChain address three challenges to ensure the poisoned documents are reliably retrieved and trusted by the LLM, even against large knowledge bases and LLM's own knowledge. Extensive experiments across six popular LLMs demonstrate that AuthChain achieves significantly higher attack success rates while maintaining superior stealthiness against RAG defense mechanisms compared to state-of-the-art baselines.
|
[
{
"version": "v1",
"created": "Thu, 15 May 2025 08:14:58 GMT"
},
{
"version": "v2",
"created": "Tue, 20 May 2025 02:50:12 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 02:28:23 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Chang",
"Zhiyuan",
""
],
[
"Li",
"Mingyang",
""
],
[
"Jia",
"Xiaojun",
""
],
[
"Wang",
"Junjie",
""
],
[
"Huang",
"Yuekai",
""
],
[
"Jiang",
"Ziyou",
""
],
[
"Liu",
"Yang",
""
],
[
"Wang",
"Qing",
""
]
] |
2505.11588
|
Ian Padilla-Gay
|
Ian Padilla-Gay, Heng-Hao Chen, Sajad Abbar, Meng-Ru Wu, Zewei Xiong
|
Flavor Equilibration of Supernova Neutrinos: Exploring the Dynamics of Slow Modes
|
15 pages, 14 figures. Minor changes, matches version accepted for publication in PRD
|
Phys. Rev. D 112, 043039 (2025)
|
10.1103/jg14-8p4l
|
SLAC-PUB-250421
|
astro-ph.HE hep-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Neutrinos experience collective flavor conversion in extreme astrophysical environments such as core-collapse supernovae (CCSNe). One manifestation of collective conversion is slow flavor conversion (SFC), which has recently attracted renewed interest owing to its ubiquity across different regions of the supernova environment. In this study, we systematically examine the evolution of kinematic decoherence in a dense neutrino gas undergoing SFC, considering lepton number asymmetries as large as $30\%$. Our findings show that the neutrino gas asymptotically evolves toward a generic state of coarse-grained flavor equilibration which is constrained by approximate lepton number conservation. The equilibration occurs within a few factors of the inverse vacuum oscillation frequency, $\omega^{-1}$, which corresponds to (anti)neutrinos reaching near flavor equipartition after a few kilometers for typical supernova neutrino energies. Notably, the quasi-steady state of the neutrino number densities can be quantitatively described by the neutrino-antineutrino number density ratio $n_{\bar{\nu}_e}/n_{\nu_e}$ alone. Such a simple estimation opens new opportunities for incorporating SFC into CCSN simulations, particularly in regions where SFC develops on scales much shorter than those of collisions.
|
[
{
"version": "v1",
"created": "Fri, 16 May 2025 18:00:00 GMT"
},
{
"version": "v2",
"created": "Thu, 14 Aug 2025 02:28:02 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Padilla-Gay",
"Ian",
""
],
[
"Chen",
"Heng-Hao",
""
],
[
"Abbar",
"Sajad",
""
],
[
"Wu",
"Meng-Ru",
""
],
[
"Xiong",
"Zewei",
""
]
] |
2505.11787
|
Rahul Deshpande
|
Atharva Mahajan, Rahul Deshpande, Taygun R. Gungor, Yvan Maciel, Ricardo Vinuesa
|
Upstream history quantification and scale-decomposed energy analysis for weak-to-strong adverse-pressure-gradient turbulent boundary layers
|
15 pages, 9 figures, accepted for publication in the International Journal of Heat and Fluid Flow
|
International Journal of Heat and Fluid Flow, Volume 117, Part A, January 2026, 110004
|
10.1016/j.ijheatfluidflow.2025.110004
| null |
physics.flu-dyn
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
The present study delineates the effects of pressure gradient history and local disequilibration on the small and large-scale energy in turbulent boundary layers (TBLs) imposed with a broad range of adverse-pressure-gradients (APG). This is made possible by analyzing four published high-fidelity APG TBL databases, which span weak to strong APGs and cover dynamic conditions ranging from near-equilibrium to strong disequilibrium. The influence of PG history on TBL statistics is quantified by the accumulated PG parameter ($\overline{\beta}$), proposed previously by Vinuesa et al. (2017) to study integral quanitites, which is compared here between cases at matched local PG strength ($\beta$), Reynolds number ($Re$) and ${\rm d}{\beta}/{{\rm d}{Re}}$ at nominally similar orders of magnitude. While the effects of local disequilibration (${\rm d}{\beta}/{{\rm d}{Re}}$) are investigated by considering TBL cases at matched $\beta$, $Re$, and fairly matched $\overline{\beta}$. It is found that $\overline{\beta}$ cannot unambiguously capture history effects when ${\rm d}{\beta}/{{\rm d}{Re}}$ levels are significantly high, as it does not account for the delayed response of the mean flow and turbulence, nor the attenuation of the pressure gradient effect with distance. In two comparisons of APG TBLs under strong non-equilibrium, the values of $\overline{\beta}$ and ${\rm d}{\beta}/{\rm d}{Re}$ expressed using Zagarola-Smits scaling were found to be consistent with the trends in mean velocity defect and Reynolds stresses noted previously for weak APG TBLs. While an increase in $\overline{\beta}$ is associated with energisation of both the small and large scales in the outer regions of APG TBLs, it affects only the large scales in the near-wall region. This confirms the ability of near-wall small scales to rapidly adjust to changes in PG strength.
|
[
{
"version": "v1",
"created": "Sat, 17 May 2025 01:56:30 GMT"
},
{
"version": "v2",
"created": "Fri, 15 Aug 2025 12:16:01 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Mahajan",
"Atharva",
""
],
[
"Deshpande",
"Rahul",
""
],
[
"Gungor",
"Taygun R.",
""
],
[
"Maciel",
"Yvan",
""
],
[
"Vinuesa",
"Ricardo",
""
]
] |
2505.12133
|
Steven Harris
|
Steven P Harris
|
On the symmetry energy expansion and the peak value of the bulk viscosity
|
v2: matches version accepted by PRC
| null | null | null |
nucl-th astro-ph.HE gr-qc
|
http://creativecommons.org/licenses/by/4.0/
|
The symmetry energy expansion is a useful way to parametrize the properties of dense matter near nuclear saturation density, and much work has been done to connect physical quantities like the neutron star radius and the core-crust transition density to the symmetry energy parameters. In this work, I connect the weak-interaction-driven bulk viscosity in neutron-proton-electron ($npe$) matter to the symmetry parameters by calculating the susceptibilities of dense matter in terms of the symmetry energy. I use this result to calculate the resonant-peak value of the bulk viscosity as a function of density, finding that it strongly depends on $L$, as does the minimum bulk-viscous dissipation timescale. Also resulting from this calculation is a formula for finding the conformal points of the zero-temperature equation of state. Finally, I determine for which values of the symmetry parameters the maximum r-mode-stable rotation frequency of an $npe$-matter neutron star is smaller than the Kepler frequency, in the high-temperature conditions where bulk viscosity is the dominant dissipation mechanism.
|
[
{
"version": "v1",
"created": "Sat, 17 May 2025 20:08:23 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 19:18:46 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Harris",
"Steven P",
""
]
] |
2505.12213
|
Huan Huang Dr.
|
Huan Huang, Hongliang Zhang, Yi Cai, Dusit Niyato, A. Lee Swindlehurst, Zhu Han
|
Simultaneously Exposing and Jamming Covert Communications via Disco Reconfigurable Intelligent Surfaces
|
This paper has been submitted for possible publication
| null | null | null |
eess.SP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Covert communications provide a stronger privacy protection than cryptography and physical-layer security (PLS). However, previous works on covert communications have implicitly assumed the validity of channel reciprocity, i.e., wireless channels remain constant or approximately constant during their coherence time. In this work, we investigate covert communications in the presence of a disco RIS (DRIS) deployed by the warden Willie, where the DRIS with random and time-varying reflective coefficients acts as a "disco ball", introducing timevarying fully-passive jamming (FPJ). Consequently, the channel reciprocity assumption no longer holds. The DRIS not only jams the covert transmissions between Alice and Bob, but also decreases the error probabilities of Willie's detections, without either Bob's channel knowledge or additional jamming power. To quantify the impact of the DRIS on covert communications, we first design a detection rule for the warden Willie in the presence of time-varying FPJ introduced by the DRIS. Then, we define the detection error probabilities, i.e., the false alarm rate (FAR) and the missed detection rate (MDR), as the monitoring performance metrics for Willie's detections, and the signal-to-jamming-plusnoise ratio (SJNR) as a communication performance metric for the covert transmissions between Alice and Bob. Based on the detection rule, we derive the detection threshold for the warden Willie to detect whether communications between Alice and Bob is ongoing, considering the time-varying DRIS-based FPJ. Moreover, we conduct theoretical analyses of the FAR and the MDR at the warden Willie, as well as SJNR at Bob, and then present unique properties of the DRIS-based FPJ in covert communications. We present numerical results to validate the derived theoretical analyses and evaluate the impact of DRIS on covert communications.
|
[
{
"version": "v1",
"created": "Sun, 18 May 2025 03:11:42 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 11:48:15 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Huang",
"Huan",
""
],
[
"Zhang",
"Hongliang",
""
],
[
"Cai",
"Yi",
""
],
[
"Niyato",
"Dusit",
""
],
[
"Swindlehurst",
"A. Lee",
""
],
[
"Han",
"Zhu",
""
]
] |
2505.12408
|
Minxu Liu
|
Minxu Liu, Donghai Guan, Chuhang Zheng, Chunwei Tian, Jie Wen, Qi Zhu
|
ViEEG: Hierarchical Visual Neural Representation for EEG Brain Decoding
|
25 pages, 17 figures
| null | null | null |
cs.CV cs.AI cs.HC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Understanding and decoding brain activity into visual representations is a fundamental challenge at the intersection of neuroscience and artificial intelligence. While EEG visual decoding has shown promise due to its non-invasive, and low-cost nature, existing methods suffer from Hierarchical Neural Encoding Neglect (HNEN)-a critical limitation where flat neural representations fail to model the brain's hierarchical visual processing hierarchy. Inspired by the hierarchical organization of visual cortex, we propose ViEEG, a neuro-We further adopt hierarchical contrastive learning for EEG-CLIP representation alignment, enabling zero-shot object recognition. Extensive experiments on the THINGS-EEG dataset demonstrate that ViEEG significantly outperforms previous methods by a large margin in both subject-dependent and subject-independent settings. Results on the THINGS-MEG dataset further confirm ViEEG's generalization to different neural modalities. Our framework not only advances the performance frontier but also sets a new paradigm for EEG brain decoding. inspired framework that addresses HNEN. ViEEG decomposes each visual stimulus into three biologically aligned components-contour, foreground object, and contextual scene-serving as anchors for a three-stream EEG encoder. These EEG features are progressively integrated via cross-attention routing, simulating cortical information flow from low-level to high-level vision.
|
[
{
"version": "v1",
"created": "Sun, 18 May 2025 13:19:08 GMT"
},
{
"version": "v2",
"created": "Sun, 25 May 2025 14:25:21 GMT"
},
{
"version": "v3",
"created": "Tue, 02 Sep 2025 09:03:26 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Liu",
"Minxu",
""
],
[
"Guan",
"Donghai",
""
],
[
"Zheng",
"Chuhang",
""
],
[
"Tian",
"Chunwei",
""
],
[
"Wen",
"Jie",
""
],
[
"Zhu",
"Qi",
""
]
] |
2505.12552
|
Lei Wang
|
Junliang Ye, Lei Wang and Md Zakir Hossain
|
FreqSelect: Frequency-Aware fMRI-to-Image Reconstruction
|
Accepted at the British Machine Vision Conference (BMVC 2025)
| null | null | null |
eess.IV cs.AI cs.CV cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Reconstructing natural images from functional magnetic resonance imaging (fMRI) data remains a core challenge in natural decoding due to the mismatch between the richness of visual stimuli and the noisy, low resolution nature of fMRI signals. While recent two-stage models, combining deep variational autoencoders (VAEs) with diffusion models, have advanced this task, they treat all spatial-frequency components of the input equally. This uniform treatment forces the model to extract meaning features and suppress irrelevant noise simultaneously, limiting its effectiveness. We introduce FreqSelect, a lightweight, adaptive module that selectively filters spatial-frequency bands before encoding. By dynamically emphasizing frequencies that are most predictive of brain activity and suppressing those that are uninformative, FreqSelect acts as a content-aware gate between image features and natural data. It integrates seamlessly into standard very deep VAE-diffusion pipelines and requires no additional supervision. Evaluated on the Natural Scenes dataset, FreqSelect consistently improves reconstruction quality across both low- and high-level metrics. Beyond performance gains, the learned frequency-selection patterns offer interpretable insights into how different visual frequencies are represented in the brain. Our method generalizes across subjects and scenes, and holds promise for extension to other neuroimaging modalities, offering a principled approach to enhancing both decoding accuracy and neuroscientific interpretability.
|
[
{
"version": "v1",
"created": "Sun, 18 May 2025 21:45:06 GMT"
},
{
"version": "v2",
"created": "Fri, 29 Aug 2025 18:38:29 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Ye",
"Junliang",
""
],
[
"Wang",
"Lei",
""
],
[
"Hossain",
"Md Zakir",
""
]
] |
2505.12620
|
Haiquan Wen
|
Haiquan Wen, Yiwei He, Zhenglin Huang, Tianxiao Li, Zihan Yu, Xingru Huang, Lu Qi, Baoyuan Wu, Xiangtai Li, Guangliang Cheng
|
BusterX: MLLM-Powered AI-Generated Video Forgery Detection and Explanation
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Advances in AI generative models facilitate super-realistic video synthesis, amplifying misinformation risks via social media and eroding trust in digital content. Several research works have explored new deepfake detection methods on AI-generated images to alleviate these risks. However, with the fast development of video generation models, such as Sora and WanX, there is currently a lack of large-scale, high-quality AI-generated video datasets for forgery detection. In addition, existing detection approaches predominantly treat the task as binary classification, lacking explainability in model decision-making and failing to provide actionable insights or guidance for the public. To address these challenges, we propose \textbf{GenBuster-200K}, a large-scale AI-generated video dataset featuring 200K high-resolution video clips, diverse latest generative techniques, and real-world scenes. We further introduce \textbf{BusterX}, a novel AI-generated video detection and explanation framework leveraging multimodal large language model (MLLM) and reinforcement learning for authenticity determination and explainable rationale. To our knowledge, GenBuster-200K is the {\it \textbf{first}} large-scale, high-quality AI-generated video dataset that incorporates the latest generative techniques for real-world scenarios. BusterX is the {\it \textbf{first}} framework to integrate MLLM with reinforcement learning for explainable AI-generated video detection. Extensive comparisons with state-of-the-art methods and ablation studies validate the effectiveness and generalizability of BusterX. The code, models, and datasets will be released.
|
[
{
"version": "v1",
"created": "Mon, 19 May 2025 02:06:43 GMT"
},
{
"version": "v2",
"created": "Wed, 21 May 2025 10:26:56 GMT"
},
{
"version": "v3",
"created": "Tue, 01 Jul 2025 19:19:43 GMT"
},
{
"version": "v4",
"created": "Thu, 31 Jul 2025 11:51:00 GMT"
},
{
"version": "v5",
"created": "Mon, 01 Sep 2025 16:38:16 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Wen",
"Haiquan",
""
],
[
"He",
"Yiwei",
""
],
[
"Huang",
"Zhenglin",
""
],
[
"Li",
"Tianxiao",
""
],
[
"Yu",
"Zihan",
""
],
[
"Huang",
"Xingru",
""
],
[
"Qi",
"Lu",
""
],
[
"Wu",
"Baoyuan",
""
],
[
"Li",
"Xiangtai",
""
],
[
"Cheng",
"Guangliang",
""
]
] |
2505.12813
|
Roberto Tauraso
|
James A. Sellers, Roberto Tauraso
|
Arithmetic properties of MacMahon-type sums of divisors: the odd case
| null |
Journal of Combinatorial Theory, Series A Volume 218, February 2026, 106105
|
10.1016/j.jcta.2025.106105
| null |
math.NT math.CO
|
http://creativecommons.org/licenses/by/4.0/
|
A century ago, P. A. MacMahon introduced two families of generating functions, $$ \sum_{1\leq n_1<n_2<\cdots<n_t}\prod_{k=1}^t\frac{q^{n_k}}{(1-q^{n_k})^2} \quad\text{ and } \sum_{\substack{1\leq n_1<n_2<\cdots<n_t\\ \text{$n_1,n_2,\dots,n_t$ odd}}}\prod_{k=1}^t\frac{q^{n_k}}{(1-q^{n_k})^2}, $$ which connect sum-of-divisors functions and integer partitions. These have recently drawn renewed attention. In particular, Amdeberhan, Andrews, and Tauraso extended the first family above by defining $$ U_t(a,q):=\sum_{1\leq n_1<n_2<\cdots<n_t}\prod_{k=1}^t\frac{q^{n_k}}{1+aq^{n_k}+q^{2n_k}} $$ for $a=0, \pm1, \pm2$ and investigated various properties, including some congruences satisfied by the coefficients of the power series representations for $U_t(a,q)$. These arithmetic aspects were subsequently expanded upon by the authors of the present work. Our goal here is to generalize the second family of generating functions, where the sums run over odd integers, and then apply similar techniques to show new infinite families of Ramanujan--like congruences for the associated power series coefficients.
|
[
{
"version": "v1",
"created": "Mon, 19 May 2025 07:43:07 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Sellers",
"James A.",
""
],
[
"Tauraso",
"Roberto",
""
]
] |
2505.12953
|
Marcello Sega
|
Pal Jedlovszky, Christoph Dellago, Marcello Sega
|
Hydrogen Bond Topology Reveals Layering of LDL-like and HDL-like Water at its Liquid/Vapor Interface
|
submitted to J Chem Phys
|
J. Chem. Phys. 163, 024511 (2025)
|
10.1063/5.0272802
| null |
cond-mat.soft physics.chem-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The discovery of high-density liquid (HDL) and low-density liquid (LDL) water has been a major success of molecular simulations, yet extending this analysis to interfacial water is challenging due to conventional order parameters assuming local homogeneity. This limitation previously prevented resolving the composition of the surface layer of the liquid/vapour interface. Here, we apply a recently introduced topological order parameter [R. Foffi and F. Sciortino, J. Phys. Chem. B 127, 378-386 (2022)] to analyze the composition of the water/vapor interface across a broad temperature range. Our results reveal that LDL-like water dominates the outermost region at all temperatures, while HDL-like water accumulates beneath it, presenting a clear layering roughly below the temperature of maximum density. This structured stratification, previously inaccessible, highlights the power of the topological order parameter in resolving interfacial molecular heterogeneity and provides new insights into the structural properties of water at interfaces.
|
[
{
"version": "v1",
"created": "Mon, 19 May 2025 10:45:33 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Jedlovszky",
"Pal",
""
],
[
"Dellago",
"Christoph",
""
],
[
"Sega",
"Marcello",
""
]
] |
2505.13259
|
Tianshi Zheng
|
Tianshi Zheng, Zheye Deng, Hong Ting Tsang, Weiqi Wang, Jiaxin Bai, Zihao Wang, Yangqiu Song
|
From Automation to Autonomy: A Survey on Large Language Models in Scientific Discovery
|
EMNLP 2025 Main
| null | null | null |
cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Large Language Models (LLMs) are catalyzing a paradigm shift in scientific discovery, evolving from task-specific automation tools into increasingly autonomous agents and fundamentally redefining research processes and human-AI collaboration. This survey systematically charts this burgeoning field, placing a central focus on the changing roles and escalating capabilities of LLMs in science. Through the lens of the scientific method, we introduce a foundational three-level taxonomy-Tool, Analyst, and Scientist-to delineate their escalating autonomy and evolving responsibilities within the research lifecycle. We further identify pivotal challenges and future research trajectories such as robotic automation, self-improvement, and ethical governance. Overall, this survey provides a conceptual architecture and strategic foresight to navigate and shape the future of AI-driven scientific discovery, fostering both rapid innovation and responsible advancement. Github Repository: https://github.com/HKUST-KnowComp/Awesome-LLM-Scientific-Discovery.
|
[
{
"version": "v1",
"created": "Mon, 19 May 2025 15:41:32 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 15:45:03 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Zheng",
"Tianshi",
""
],
[
"Deng",
"Zheye",
""
],
[
"Tsang",
"Hong Ting",
""
],
[
"Wang",
"Weiqi",
""
],
[
"Bai",
"Jiaxin",
""
],
[
"Wang",
"Zihao",
""
],
[
"Song",
"Yangqiu",
""
]
] |
2505.13430
|
Sifeng Shang
|
Sifeng Shang, Jiayi Zhou, Chenyu Lin, Minxian Li, Kaiyang Zhou
|
Fine-tuning Quantized Neural Networks with Zeroth-order Optimization
| null | null | null | null |
cs.LG cs.CL cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
As the size of large language models grows exponentially, GPU memory has become a bottleneck for adapting these models to downstream tasks. In this paper, we aim to push the limits of memory-efficient training by minimizing memory usage on model weights, gradients, and optimizer states, within a unified framework. Our idea is to eliminate both gradients and optimizer states using zeroth-order optimization, which approximates gradients by perturbing weights during forward passes to identify gradient directions. To minimize memory usage on weights, we employ model quantization, e.g., converting from bfloat16 to int4. However, directly applying zeroth-order optimization to quantized weights is infeasible due to the precision gap between discrete weights and continuous gradients, which would otherwise require de-quantization and re-quantization. To overcome this challenge, we propose Quantized Zeroth-order Optimization (QZO), a simple yet effective approach that perturbs the continuous quantization scale for gradient estimation and uses a directional derivative clipping method to stabilize training. QZO is orthogonal to both scalar-based and codebook-based post-training quantization methods. Compared to full-parameter fine-tuning in 16 bits, QZO can reduce the total memory cost by more than 18$\times$ for 4-bit LLMs, and enables fine-tuning Llama-2-13B within a single 24GB GPU. Code will be released publicly.
|
[
{
"version": "v1",
"created": "Mon, 19 May 2025 17:55:15 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 11:50:29 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Shang",
"Sifeng",
""
],
[
"Zhou",
"Jiayi",
""
],
[
"Lin",
"Chenyu",
""
],
[
"Li",
"Minxian",
""
],
[
"Zhou",
"Kaiyang",
""
]
] |
2505.13578
|
Mikhail Osipov
|
Mikhail Osipov
|
Symmetry-Breaking Descent for Invariant Cost Functionals
|
20 pages, 4 appendices
| null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
We study the problem of reducing a task cost functional $W : H^s(M) \to \mathbb{R}$, not assumed continuous or differentiable, defined over Sobolev-class signals $S \in H^s(M) $, in the presence of a global symmetry group $G \subset \mathrm{Diff}(M)$. The group acts on signals by pullback, and the cost $W$ is invariant under this action. Such scenarios arise in machine learning and related optimization tasks, where performance metrics may be discontinuous or model-internal.
We propose a variational method that exploits the symmetry structure to construct explicit deformations of the input signal. A deformation control field $ \phi: M \to \mathbb R^d$, obtained by minimizing an auxiliary energy functional, induces a flow that generically lies in the normal space (with respect to the $L^2$ inner product) to the $G$-orbit of $S$, and hence is a natural candidate to cross the decision boundary of the $G $-invariant cost.
We analyze two variants of the coupling term: (1) purely geometric, independent of $W$, and (2) weakly coupled to $W$. Under mild conditions, we show that symmetry-breaking deformations of the signal can reduce the cost.
Our approach requires no gradient backpropagation or training labels and operates entirely at test time. It provides a principled tool for optimizing discontinuous invariant cost functionals via Lie-algebraic variational flows.
|
[
{
"version": "v1",
"created": "Mon, 19 May 2025 15:06:31 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 17:41:32 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Osipov",
"Mikhail",
""
]
] |
2505.13834
|
Zhi Su
|
Zhi Su, Yuman Gao, Emily Lukas, Yunfei Li, Jiaze Cai, Faris Tulbah, Fei Gao, Chao Yu, Zhongyu Li, Yi Wu, Koushil Sreenath
|
Toward Real-World Cooperative and Competitive Soccer with Quadrupedal Robot Teams
|
11 pages, 12 figures, CoRL 2025
| null | null | null |
cs.RO cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Achieving coordinated teamwork among legged robots requires both fine-grained locomotion control and long-horizon strategic decision-making. Robot soccer offers a compelling testbed for this challenge, combining dynamic, competitive, and multi-agent interactions. In this work, we present a hierarchical multi-agent reinforcement learning (MARL) framework that enables fully autonomous and decentralized quadruped robot soccer. First, a set of highly dynamic low-level skills is trained for legged locomotion and ball manipulation, such as walking, dribbling, and kicking. On top of these, a high-level strategic planning policy is trained with Multi-Agent Proximal Policy Optimization (MAPPO) via Fictitious Self-Play (FSP). This learning framework allows agents to adapt to diverse opponent strategies and gives rise to sophisticated team behaviors, including coordinated passing, interception, and dynamic role allocation. With an extensive ablation study, the proposed learning method shows significant advantages in the cooperative and competitive multi-agent soccer game. We deploy the learned policies to real quadruped robots relying solely on onboard proprioception and decentralized localization, with the resulting system supporting autonomous robot-robot and robot-human soccer matches on indoor and outdoor soccer courts.
|
[
{
"version": "v1",
"created": "Tue, 20 May 2025 02:20:54 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 02:08:59 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Su",
"Zhi",
""
],
[
"Gao",
"Yuman",
""
],
[
"Lukas",
"Emily",
""
],
[
"Li",
"Yunfei",
""
],
[
"Cai",
"Jiaze",
""
],
[
"Tulbah",
"Faris",
""
],
[
"Gao",
"Fei",
""
],
[
"Yu",
"Chao",
""
],
[
"Li",
"Zhongyu",
""
],
[
"Wu",
"Yi",
""
],
[
"Sreenath",
"Koushil",
""
]
] |
2505.14077
|
Freija Beirnaert
|
Freija Beirnaert, Archisman Ghosh, Gergely D\'alya
|
A Hubble constant estimation with dark standard sirens and galaxy cluster catalogues
| null | null | null | null |
astro-ph.CO gr-qc
|
http://creativecommons.org/licenses/by/4.0/
|
In this paper, we explore the possibility of using galaxy cluster catalogues to provide redshift support for a gravitational-wave dark standard siren measurement of the Hubble constant $H_0$. We adapt the cosmology inference pipeline gwcosmo to handle galaxy cluster catalogues. Together with binary black holes from the GWTC-3, we use galaxy cluster data from the PSZ2 and the eRASS catalogues. With these catalogues, we obtain $H_0 = 77^{+10}_{-10}$ and $81^{+8}_{-8}\, \text{km}\, \text{s}^{-1}\, \text{Mpc}^{-1}$ respectively, which demonstrates improvements on precision by factors of 10% and 38% respectively over the traditional galaxy catalogue result. This exploratory work paves the way towards precise and accurate cosmography making use of distant compact binary mergers from upcoming observing runs of the LIGO-Virgo-KAGRA detector network and future gravitational-wave observatories.
|
[
{
"version": "v1",
"created": "Tue, 20 May 2025 08:34:23 GMT"
},
{
"version": "v2",
"created": "Fri, 29 Aug 2025 20:33:18 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Beirnaert",
"Freija",
""
],
[
"Ghosh",
"Archisman",
""
],
[
"Dálya",
"Gergely",
""
]
] |
2505.14382
|
Jorge Peixinho
|
Vin\'icius Pereira da S. Oliveira, Danilo S. Borges, Erick M. Franklin, Jorge Peixinho
|
Role of Friction on the Formation of Confined Granular Structures
|
7 pages, 7 figures
| null | null | null |
cond-mat.soft physics.flu-dyn
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Unstable systems of fluidized grains in a very-narrow vertical tube can auto-defluidize after some time, the settling particles forming either a glass- or crystal-like structure. We carried out experiments using different polymer spheres, of known friction and roughness, fluidized in water. A diagram was obtained for the \RefereeA{shell-settled} particles when the coefficient of friction is of the order of 0.1, and their structure is characterized through an analysis of the nearest neighbors' angles. We show that the level of velocity fluctuations is higher for the high friction material, and that, once defluidized, the decrease in the coefficient of sliding friction leads to more organized (crystal-like) structures, while those with higher friction coefficients are amorphous (glass-like structures). Our results bring new insights for understanding the formation of glass- and crystal-like structures based on the material surface properties.
|
[
{
"version": "v1",
"created": "Tue, 20 May 2025 14:03:48 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 20:19:41 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Oliveira",
"Vinícius Pereira da S.",
""
],
[
"Borges",
"Danilo S.",
""
],
[
"Franklin",
"Erick M.",
""
],
[
"Peixinho",
"Jorge",
""
]
] |
2505.14590
|
Huihao Jing
|
Huihao Jing, Haoran Li, Wenbin Hu, Qi Hu, Heli Xu, Tianshu Chu, Peizhao Hu, Yangqiu Song
|
MCIP: Protecting MCP Safety via Model Contextual Integrity Protocol
|
17 pages
| null | null | null |
cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
As Model Context Protocol (MCP) introduces an easy-to-use ecosystem for users and developers, it also brings underexplored safety risks. Its decentralized architecture, which separates clients and servers, poses unique challenges for systematic safety analysis. This paper proposes a novel framework to enhance MCP safety. Guided by the MAESTRO framework, we first analyze the missing safety mechanisms in MCP, and based on this analysis, we propose the Model Contextual Integrity Protocol (MCIP), a refined version of MCP that addresses these gaps. Next, we develop a fine-grained taxonomy that captures a diverse range of unsafe behaviors observed in MCP scenarios. Building on this taxonomy, we develop benchmark and training data that support the evaluation and improvement of LLMs' capabilities in identifying safety risks within MCP interactions. Leveraging the proposed benchmark and training data, we conduct extensive experiments on state-of-the-art LLMs. The results highlight LLMs' vulnerabilities in MCP interactions and demonstrate that our approach substantially improves their safety performance.
|
[
{
"version": "v1",
"created": "Tue, 20 May 2025 16:41:45 GMT"
},
{
"version": "v2",
"created": "Wed, 21 May 2025 14:48:40 GMT"
},
{
"version": "v3",
"created": "Fri, 30 May 2025 06:55:41 GMT"
},
{
"version": "v4",
"created": "Wed, 04 Jun 2025 06:21:46 GMT"
},
{
"version": "v5",
"created": "Tue, 02 Sep 2025 13:04:46 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Jing",
"Huihao",
""
],
[
"Li",
"Haoran",
""
],
[
"Hu",
"Wenbin",
""
],
[
"Hu",
"Qi",
""
],
[
"Xu",
"Heli",
""
],
[
"Chu",
"Tianshu",
""
],
[
"Hu",
"Peizhao",
""
],
[
"Song",
"Yangqiu",
""
]
] |
2505.14755
|
Ang Li
|
Mingyang Wang, Peng Liu, Jianping Yuan, Ang Li, Youli Tuo, Shijun Dang, Weihua Wang, Mingyu Ge, Xia Zhou, and Na Wang
|
Discovery of the anti-glitch in PSR J1835$-$1106
|
7 pages, 2 figures, 3 tables, version accepted for publication in ApJL
| null | null | null |
astro-ph.HE astro-ph.SR nucl-th
|
http://creativecommons.org/licenses/by/4.0/
|
We report the detection of an anti-glitch with a fractional frequency change of $\Delta\nu/\nu=-3.46(6)\times10^{-9}$ in the rotation-powered pulsar PSR J1835$-$1106 at MJD 55813, based on timing observations collected with the Nanshan 26-m and Parkes 64-m radio telescopes from January 2000 to July 2022. A comparison of the average pulse profiles within $\pm300$ d of the event reveals no significant morphological changes. We also estimate the angular velocity lag between the normal and superfluid components at the time of the glitch, showing that one of the superfluid glitch models is incompatible with PSR J1835$-$1106 due to its insufficient spin-down rate and angular velocity lag. The wind braking scenario offers a viable alternative, consistent with the observed spin-down behavior, glitch amplitude, and post-glitch recovery. High-cadence, high-sensitivity monitoring of similar events is essential to distinguish between internal (superfluid) and external (wind-related) glitch mechanisms.
|
[
{
"version": "v1",
"created": "Tue, 20 May 2025 15:24:14 GMT"
},
{
"version": "v2",
"created": "Sun, 06 Jul 2025 06:48:47 GMT"
},
{
"version": "v3",
"created": "Sun, 31 Aug 2025 15:00:49 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Wang",
"Mingyang",
""
],
[
"Liu",
"Peng",
""
],
[
"Yuan",
"Jianping",
""
],
[
"Li",
"Ang",
""
],
[
"Tuo",
"Youli",
""
],
[
"Dang",
"Shijun",
""
],
[
"Wang",
"Weihua",
""
],
[
"Ge",
"Mingyu",
""
],
[
"Zhou",
"Xia",
""
],
[
"Wang",
"Na",
""
]
] |
2505.15327
|
Jeremy Singer
|
Jeremy Singer and Steve Draper
|
Let's Take Esoteric Programming Languages Seriously
|
13 pages, 7 figures
| null |
10.1145/3759429.3762632
| null |
cs.PL
|
http://creativecommons.org/licenses/by/4.0/
|
Esoteric programming languages are challenging to learn, but their unusual features and constraints may serve to improve programming ability. From languages designed to be intentionally obtuse (e.g. INTERCAL) to others targeting artistic expression (e.g. Piet) or exploring the nature of computation (e.g. Fractan), there is rich variety in the realm of esoteric programming languages. This essay examines the counterintuitive appeal of esoteric languages and seeks to analyse reasons for this popularity. We will explore why people are attracted to esoteric languages in terms of (a) program comprehension and construction, as well as (b) language design and implementation. Our assertion is that esoteric languages can improve general PL awareness, at the same time as enabling the esoteric programmer to impress their peers with obscure knowledge. We will also consider pedagogic principles and the use of AI, in relation to esoteric languages. Emerging from the specific discussion, we identify a general set of 'good' reasons for designing new programming languages. It may not be possible to be exhaustive on this topic, and it is certain we have not achieved that goal here. However we believe our most important contribution is to draw attention to the varied and often implicit motivations involved in programming language design.
|
[
{
"version": "v1",
"created": "Wed, 21 May 2025 10:02:07 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 13:20:49 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Singer",
"Jeremy",
""
],
[
"Draper",
"Steve",
""
]
] |
2505.15653
|
Yong See Foo Mr
|
Yong See Foo, Adriana Zanca, Jennifer A. Flegg, Ivo Siekmann
|
Quantifying structural uncertainty in chemical reaction network inference
|
36 pages, 12 figures
| null | null | null |
stat.ME q-bio.QM
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Dynamical systems in chemistry and biology are complex, and one often does not have comprehensive knowledge about the interactions involved. Chemical reaction network (CRN) inference aims to identify, from observing species concentrations over time, the unknown reactions between the species. Existing approaches largely focus on identifying a single, most likely CRN, without addressing uncertainty about the network structure. However, it is important to quantify structural uncertainty to have confidence in our inference and predictions. In this work, we do so by inferring an approximate posterior distribution over CRN structures. This is done by keeping a large set of suboptimal solutions found through sparse optimisation, in contrast to existing optimisation approaches which discard suboptimal solutions. We find that inducing reaction sparsity with nonconvex penalty functions results in more parsimonious CRNs compared to the popular lasso regularisation. In a real-data example where multiple CRNs have been previously suggested, our method simultaneously recovers reactions proposed from different literature. Our emphasis on network-level probabilities enables a novel, hierarchical representation of structural ambiguities in the space of CRNs. This readily translates into alternative reaction pathways suggested by the available data, thus guiding the efforts of future experimental design.
|
[
{
"version": "v1",
"created": "Wed, 21 May 2025 15:31:00 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 05:29:19 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Foo",
"Yong See",
""
],
[
"Zanca",
"Adriana",
""
],
[
"Flegg",
"Jennifer A.",
""
],
[
"Siekmann",
"Ivo",
""
]
] |
2505.15695
|
Ryang Heo
|
Ryang Heo, Yongsik Seo, Junseong Lee, Dongha Lee
|
Can Large Language Models be Effective Online Opinion Miners?
|
Accepted to EMNLP 2025 Main
| null | null | null |
cs.CL
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
The surge of user-generated online content presents a wealth of insights into customer preferences and market trends. However, the highly diverse, complex, and context-rich nature of such contents poses significant challenges to traditional opinion mining approaches. To address this, we introduce Online Opinion Mining Benchmark (OOMB), a novel dataset and evaluation protocol designed to assess the ability of large language models (LLMs) to mine opinions effectively from diverse and intricate online environments. OOMB provides extensive (entity, feature, opinion) tuple annotations and a comprehensive opinion-centric summary that highlights key opinion topics within each content, thereby enabling the evaluation of both the extractive and abstractive capabilities of models. Through our proposed benchmark, we conduct a comprehensive analysis of which aspects remain challenging and where LLMs exhibit adaptability, to explore whether they can effectively serve as opinion miners in realistic online scenarios. This study lays the foundation for LLM-based opinion mining and discusses directions for future research in this field.
|
[
{
"version": "v1",
"created": "Wed, 21 May 2025 16:09:44 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 15:48:33 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Heo",
"Ryang",
""
],
[
"Seo",
"Yongsik",
""
],
[
"Lee",
"Junseong",
""
],
[
"Lee",
"Dongha",
""
]
] |
2505.15737
|
Zhuodong Jiang
|
Zhuodong Jiang, Haoran Wang, Guoxi Huang, Brett Seymour, Nantheera Anantrasirichai
|
RUSplatting: Robust 3D Gaussian Splatting for Sparse-View Underwater Scene Reconstruction
|
Accepted by BMVC 2025
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Reconstructing high-fidelity underwater scenes remains a challenging task due to light absorption, scattering, and limited visibility inherent in aquatic environments. This paper presents an enhanced Gaussian Splatting-based framework that improves both the visual quality and geometric accuracy of deep underwater rendering. We propose decoupled learning for RGB channels, guided by the physics of underwater attenuation, to enable more accurate colour restoration. To address sparse-view limitations and improve view consistency, we introduce a frame interpolation strategy with a novel adaptive weighting scheme. Additionally, we introduce a new loss function aimed at reducing noise while preserving edges, which is essential for deep-sea content. We also release a newly collected dataset, Submerged3D, captured specifically in deep-sea environments. Experimental results demonstrate that our framework consistently outperforms state-of-the-art methods with PSNR gains up to 1.90dB, delivering superior perceptual quality and robustness, and offering promising directions for marine robotics and underwater visual analytics. The code of RUSplatting is available at https://github.com/theflash987/RUSplatting and the dataset Submerged3D can be downloaded at https://zenodo.org/records/15482420.
|
[
{
"version": "v1",
"created": "Wed, 21 May 2025 16:42:15 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 19:24:08 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Jiang",
"Zhuodong",
""
],
[
"Wang",
"Haoran",
""
],
[
"Huang",
"Guoxi",
""
],
[
"Seymour",
"Brett",
""
],
[
"Anantrasirichai",
"Nantheera",
""
]
] |
2505.15883
|
Jakob Moritz
|
Jakob Moritz
|
$G_2$-manifolds from Diophantine equations
|
49 pages, 1 figure; v2: added footnote and references
| null | null | null |
hep-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We argue that \emph{perturbatively flat vacua} (PFVs) introduced in \cite{Demirtas:2019sip} are dual to M-theory compactifications on $G_2$-manifolds, enabling the enumeration of novel $G_2$-manifolds via solutions to Diophantine equations in type IIB flux quanta. Independently, we show that warping corrections to the effective action of type IIB flux vacua grow parametrically at large complex structure, and we demonstrate that these corrections can nonetheless be captured by a classical geometric computation in M-theory.
|
[
{
"version": "v1",
"created": "Wed, 21 May 2025 18:00:00 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 19:40:04 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Moritz",
"Jakob",
""
]
] |
2505.16131
|
Thomas Vandermeulen
|
Nathan Brady, David Tennyson, Thomas Vandermeulen
|
Machine Learning the 6d Supergravity Landscape
|
50 pages; code and data available at https://github.com/nait400/ML-6d-sugra-landscape; v2: references added
| null | null | null |
hep-th cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this paper, we apply both supervised and unsupervised machine learning algorithms to the study of the string landscape and swampland in 6-dimensions. Our data are the (almost) anomaly-free 6-dimensional $\mathcal{N} = (1,0)$ supergravity models, characterised by the Gram matrix of anomaly coefficients. Our work demonstrates the ability of machine learning algorithms to efficiently learn highly complex features of the landscape and swampland. Employing an autoencoder for unsupervised learning, we provide an auto-classification of these models by compressing the Gram matrix data to 2-dimensions. Through compression, similar models cluster together, and we identify prominent features of these clusters. The autoencoder also identifies outlier models which are difficult to reconstruct. One of these outliers proves to be incredibly difficult to combine with other models such that the $\text{tr}R^{4}$ anomaly vanishes, making its presence in the landscape extremely rare. Further, we utilise supervised learning to build two classifiers predicting (1) model consistency under probe string insertion (precision: 0.78, predicting consistency for 214,837 models with reasonable certainty) and (2) inconsistency under anomaly inflow (precision: 0.91, predicting inconsistency for 1,909,359 models). Notably, projecting these predictions onto the autoencoder's 2-dimensional latent layer shows consistent models clustering together, further indicating that the autoencoder has learnt interesting and complex features of the set of models and potentially offers a novel approach to mapping the landscape and swampland of 6-dimensional supergravity theories.
|
[
{
"version": "v1",
"created": "Thu, 22 May 2025 02:16:55 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 02:14:43 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Brady",
"Nathan",
""
],
[
"Tennyson",
"David",
""
],
[
"Vandermeulen",
"Thomas",
""
]
] |
2505.16172
|
Abhay Kumara Sri Krishna Nandiraju
|
Abhay Kumara Sri Krishna Nandiraju, Gondy Leroy, David Kauchak, and Arif Ahmed
|
Automated Feedback Loops to Protect Text Simplification with Generative AI from Information Loss
| null | null |
10.1007/978-3-032-00071-2_31
| null |
cs.CL cs.AI
|
http://creativecommons.org/licenses/by-sa/4.0/
|
Understanding health information is essential in achieving and maintaining a healthy life. We focus on simplifying health information for better understanding. With the availability of generative AI, the simplification process has become efficient and of reasonable quality, however, the algorithms remove information that may be crucial for comprehension. In this study, we compare generative AI to detect missing information in simplified text, evaluate its importance, and fix the text with the missing information. We collected 50 health information texts and simplified them using gpt-4-0613. We compare five approaches to identify missing elements and regenerate the text by inserting the missing elements. These five approaches involve adding missing entities and missing words in various ways: 1) adding all the missing entities, 2) adding all missing words, 3) adding the top-3 entities ranked by gpt-4-0613, and 4, 5) serving as controls for comparison, adding randomly chosen entities. We use cosine similarity and ROUGE scores to evaluate the semantic similarity and content overlap between the original, simplified, and reconstructed simplified text. We do this for both summaries and full text. Overall, we find that adding missing entities improves the text. Adding all the missing entities resulted in better text regeneration, which was better than adding the top-ranked entities or words, or random words. Current tools can identify these entities, but are not valuable in ranking them.
|
[
{
"version": "v1",
"created": "Thu, 22 May 2025 03:19:49 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Nandiraju",
"Abhay Kumara Sri Krishna",
""
],
[
"Leroy",
"Gondy",
""
],
[
"Kauchak",
"David",
""
],
[
"Ahmed",
"Arif",
""
]
] |
2505.16397
|
Koki Nagakura
|
Koki Nagakura, Tatsuki Fushimi, Ayaka Tsutsui, and Yoichi Ochiai
|
Dynamic Caustics by Ultrasonically Modulated Liquid Surface
| null | null |
10.1038/s41598-025-16190-3
| null |
cs.GR cs.HC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper presents a method for generating dynamic caustic patterns by utilising dual-optimised holographic fields with Phased Array Transducer (PAT). Building on previous research in static caustic optimisation and ultrasonic manipulation, this approach employs computational techniques to dynamically shape fluid surfaces, thereby creating controllable and real-time caustic images. The system employs a Digital Twin framework, which enables iterative feedback and refinement, thereby improving the accuracy and quality of the caustic patterns produced. This paper extends the foundational work in caustic generation by integrating liquid surfaces as refractive media. This concept has previously been explored in simulations but not fully realised in practical applications. The utilisation of ultrasound to directly manipulate these surfaces enables the generation of dynamic caustics with a high degree of flexibility. The Digital Twin approach further enhances this process by allowing for precise adjustments and optimisation based on real-time feedback. Experimental results demonstrate the technique's capacity to generate continuous animations and complex caustic patterns at high frequencies. Although there are limitations in contrast and resolution compared to solid-surface methods, this approach offers advantages in terms of real-time adaptability and scalability. This technique has the potential to be applied in a number of areas, including interactive displays, artistic installations and educational tools. This research builds upon the work of previous researchers in the fields of caustics optimisation, ultrasonic manipulation, and computational displays. Future research will concentrate on enhancing the resolution and intricacy of the generated patterns.
|
[
{
"version": "v1",
"created": "Thu, 22 May 2025 08:49:14 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Nagakura",
"Koki",
""
],
[
"Fushimi",
"Tatsuki",
""
],
[
"Tsutsui",
"Ayaka",
""
],
[
"Ochiai",
"Yoichi",
""
]
] |
2505.16647
|
Sushant Gautam
|
Sushant Gautam, Michael A. Riegler, P{\aa}l Halvorsen
|
Point, Detect, Count: Multi-Task Medical Image Understanding with Instruction-Tuned Vision-Language Models
|
Accepted as a full paper at the 38th IEEE International Symposium on Computer-Based Medical Systems (CBMS) 2025
| null |
10.1109/CBMS65348.2025.00090
| null |
cs.CV cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
We investigate fine-tuning Vision-Language Models (VLMs) for multi-task medical image understanding, focusing on detection, localization, and counting of findings in medical images. Our objective is to evaluate whether instruction-tuned VLMs can simultaneously improve these tasks, with the goal of enhancing diagnostic accuracy and efficiency. Using MedMultiPoints, a multimodal dataset with annotations from endoscopy (polyps and instruments) and microscopy (sperm cells), we reformulate each task into instruction-based prompts suitable for vision-language reasoning. We fine-tune Qwen2.5-VL-7B-Instruct using Low-Rank Adaptation (LoRA) across multiple task combinations. Results show that multi-task training improves robustness and accuracy. For example, it reduces the Count Mean Absolute Error (MAE) and increases Matching Accuracy in the Counting + Pointing task. However, trade-offs emerge, such as more zero-case point predictions, indicating reduced reliability in edge cases despite overall performance gains. Our study highlights the potential of adapting general-purpose VLMs to specialized medical tasks via prompt-driven fine-tuning. This approach mirrors clinical workflows, where radiologists simultaneously localize, count, and describe findings - demonstrating how VLMs can learn composite diagnostic reasoning patterns. The model produces interpretable, structured outputs, offering a promising step toward explainable and versatile medical AI. Code, model weights, and scripts will be released for reproducibility at https://github.com/simula/PointDetectCount.
|
[
{
"version": "v1",
"created": "Thu, 22 May 2025 13:18:44 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Gautam",
"Sushant",
""
],
[
"Riegler",
"Michael A.",
""
],
[
"Halvorsen",
"Pål",
""
]
] |
2505.16704
|
Jo\'as Ven\^ancio
|
Jo\'as Ven\^ancio, Herondy Mota, Azadeh Mohammadi
|
Finite temperature Casimir effect for a spinor field in cosmic dispiration spacetime
|
21 pages, 4 figures
|
JHEP 09, 007 (2025)
|
10.1007/JHEP09(2025)007
| null |
hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
This study explores the finite temperature Casimir effect for a massive spinor field in cosmic dispiration spacetime, formed by the combination of a cosmic string and a screw dislocation using the generalized zeta function regularization method. First, we examine the cosmic string spacetime with a quasi-antiperiodic boundary condition, where the Casimir energy and its corrections depend on two nonzero heat kernel coefficients, one associated with the Euclidean divergence and the other with the nontrivial topology, both vanishing when renormalized. Interestingly, for specific choice of parameters the quasi-antiperiodicity effect can entirely cancel out the topological contribution, leaving only the Euclidean divergence. We then extend this analysis to cosmic dispiration spacetime. This configuration alters the spacetime topology, modifying the structure of the heat kernel coefficient related to the new nontrivial topology. In this case, the renormalized Casimir energy density can take positive or negative values and decreases exponentially as the field mass increases. Additionally, we examine the asymptotic behavior of the renormalized temperature correction term in the massless regime, showing that the spinor vacuum free energy vanishes at very high temperatures. At very low temperatures, it is dominated by the zero-temperature Casimir energy density.
|
[
{
"version": "v1",
"created": "Thu, 22 May 2025 14:05:21 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Venâncio",
"Joás",
""
],
[
"Mota",
"Herondy",
""
],
[
"Mohammadi",
"Azadeh",
""
]
] |
2505.16959
|
Alessandro Favero
|
Alessandro Favero, Antonio Sclocchi, Matthieu Wyart
|
Bigger Isn't Always Memorizing: Early Stopping Overparameterized Diffusion Models
| null | null | null | null |
cs.LG stat.ML
|
http://creativecommons.org/licenses/by/4.0/
|
Diffusion probabilistic models have become a cornerstone of modern generative AI, yet the mechanisms underlying their generalization remain poorly understood. In fact, if these models were perfectly minimizing their training loss, they would just generate data belonging to their training set, i.e., memorize, as empirically found in the overparameterized regime. We revisit this view by showing that, in highly overparameterized diffusion models, generalization in natural data domains is progressively achieved during training before the onset of memorization. Our results, ranging from image to language diffusion models, systematically support the empirical law that memorization time is proportional to the dataset size. Generalization vs. memorization is then best understood as a competition between time scales. We show that this phenomenology is recovered in diffusion models learning a simple probabilistic context-free grammar with random rules, where generalization corresponds to the hierarchical acquisition of deeper grammar rules as training time grows, and the generalization cost of early stopping can be characterized. We summarize these results in a phase diagram. Overall, our results support that a principled early-stopping criterion - scaling with dataset size - can effectively optimize generalization while avoiding memorization, with direct implications for hyperparameter transfer and privacy-sensitive applications.
|
[
{
"version": "v1",
"created": "Thu, 22 May 2025 17:40:08 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 20:18:07 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Favero",
"Alessandro",
""
],
[
"Sclocchi",
"Antonio",
""
],
[
"Wyart",
"Matthieu",
""
]
] |
2505.17117
|
Chen Shani
|
Chen Shani, Liron Soffer, Dan Jurafsky, Yann LeCun, Ravid Shwartz-Ziv
|
From Tokens to Thoughts: How LLMs and Humans Trade Compression for Meaning
| null | null | null | null |
cs.CL cs.AI cs.IT math.IT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Humans organize knowledge into compact categories through semantic compression by mapping diverse instances to abstract representations while preserving meaning (e.g., robin and blue jay are both birds; most birds can fly). These concepts reflect a trade-off between expressive fidelity and representational simplicity. Large Language Models (LLMs) demonstrate remarkable linguistic abilities, yet whether their internal representations strike a human-like trade-off between compression and semantic fidelity is unclear. We introduce a novel information-theoretic framework, drawing from Rate-Distortion Theory and the Information Bottleneck principle, to quantitatively compare these strategies. Analyzing token embeddings from a diverse suite of LLMs against seminal human categorization benchmarks, we uncover key divergences. While LLMs form broad conceptual categories that align with human judgment, they struggle to capture the fine-grained semantic distinctions crucial for human understanding. More fundamentally, LLMs demonstrate a strong bias towards aggressive statistical compression, whereas human conceptual systems appear to prioritize adaptive nuance and contextual richness, even if this results in lower compressional efficiency by our measures. These findings illuminate critical differences between current AI and human cognitive architectures, guiding pathways toward LLMs with more human-aligned conceptual representations.
|
[
{
"version": "v1",
"created": "Wed, 21 May 2025 16:29:00 GMT"
},
{
"version": "v2",
"created": "Mon, 26 May 2025 21:13:36 GMT"
},
{
"version": "v3",
"created": "Mon, 30 Jun 2025 21:22:39 GMT"
},
{
"version": "v4",
"created": "Sun, 31 Aug 2025 19:56:10 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Shani",
"Chen",
""
],
[
"Soffer",
"Liron",
""
],
[
"Jurafsky",
"Dan",
""
],
[
"LeCun",
"Yann",
""
],
[
"Shwartz-Ziv",
"Ravid",
""
]
] |
2505.17174
|
Chang Liu
|
Chang Liu, Adam A. Miller, Joshua S. Bloom, Robert A. Knop, Peter E. Nugent
|
A Morphological Model to Separate Resolved-Unresolved Sources in the DESI Legacy Surveys: Application in the LS4 Alert Stream
|
26 pages, 10 figures, 9 tables. Updated to accepted version (PASP)
|
PASP 137 084501 (2025)
|
10.1088/1538-3873/adf7db
| null |
astro-ph.IM
|
http://creativecommons.org/licenses/by/4.0/
|
Separating resolved and unresolved sources in large imaging surveys is a fundamental step to enable downstream science, such as searching for extragalactic transients in wide-field time-domain surveys. Here we present our method to effectively separate point sources from the resolved, extended sources in the Dark Energy Spectroscopic Instrument (DESI) Legacy Surveys (LS). We develop a supervised machine-learning model based on the Gradient Boosting algorithm $\texttt{XGBoost}$. The features input to the model are purely morphological and are derived from the tabulated LS data products. We train the model using $\sim$$2\times10^5$ LS sources in the COSMOS field with HST morphological labels and evaluate the model performance on LS sources with spectroscopic classification from the DESI Data Release 1 ($\sim$$2\times10^7$ objects) and the Sloan Digital Sky Survey Data Release 17 ($\sim$$3\times10^6$ objects), as well as on $\sim$$2\times10^8$ Gaia stars. A significant fraction of LS sources are not observed in every LS filter, and we therefore build a ''Hybrid'' model as a linear combination of two \texttt{XGBoost} models, each containing features combining aperture flux measurements from the ''blue'' ($gr$) and ''red'' ($iz$) filters. The Hybrid model shows a reasonable balance between sensitivity and robustness, and achieves higher accuracy and flexibility compared to the LS morphological typing. With the Hybrid model, we provide classification scores for $\sim$$3\times10^9$ LS sources, making this the largest ever machine-learning catalog separating resolved and unresolved sources. The catalog has been incorporated into the real-time pipeline of the La Silla Schmidt Southern Survey (LS4), enabling the identification of extragalactic transients within the LS4 alert stream.
|
[
{
"version": "v1",
"created": "Thu, 22 May 2025 18:00:00 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 23:29:30 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Liu",
"Chang",
""
],
[
"Miller",
"Adam A.",
""
],
[
"Bloom",
"Joshua S.",
""
],
[
"Knop",
"Robert A.",
""
],
[
"Nugent",
"Peter E.",
""
]
] |
2505.17188
|
Soumak Maitra
|
Soumak Maitra (TIFR), Girish Kulkarni, Shikhar Asthana, James S. Bolton, Martin G. Haehnelt and Laura Keating
|
The Lyman-$\alpha$ emitter bispectrum as a probe of reionization morphology
|
22 pages, 7 figures, 4 tables
|
Mon Not R Astron Soc (2025) 486-507
|
10.1093/mnras/staf1262
| null |
astro-ph.CO astro-ph.GA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Ly$\alpha$ emitters (LAEs) have now been discovered out to redshift $z=13$, and are valuable probes of the reionization history at redshifts beyond the reach of other currently available tracers. Most inferences of the neutral hydrogen fraction from LAE observations rely on one-point and two-point statistics like the luminosity function and the power spectrum. We present here an analysis of the bispectrum of high-redshift LAEs and demonstrate its sensitivity to the Epoch of Reionization. We use the Sherwood-Relics suite of cosmological hydrodynamical simulations post-processed with the GPU-based radiative transfer code ATON-HE to generate realistic LAE mock catalogues for a wide range of reionization models, varying the ionization history and the source populations, including contributions of AGN to hydrogen reionization. We demonstrate that the bispectrum provides greater sensitivity than the power spectrum to both the timing and spatial morphology of reionization. Using reduced-$\chi^2$ analysis we further show that the bispectrum also responds more strongly to variations in source population and AGN contribution, apparently more efficiently capturing morphological features missed by two-point statistics. The redshift evolution of the bispectrum reflects the increased clustering of ionizing sources at earlier epochs. The sensitivity of the bispectrum to peculiar velocities underscores the importance of velocity corrections in comparisons to observations. Our findings demonstrate that the LAE bispectrum is a powerful higher-order statistic for probing reionization through current and future LAE surveys using telescopes such as Subaru and JWST.
|
[
{
"version": "v1",
"created": "Thu, 22 May 2025 18:00:09 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 15:32:59 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Maitra",
"Soumak",
"",
"TIFR"
],
[
"Kulkarni",
"Girish",
""
],
[
"Asthana",
"Shikhar",
""
],
[
"Bolton",
"James S.",
""
],
[
"Haehnelt",
"Martin G.",
""
],
[
"Keating",
"Laura",
""
]
] |
2505.17886
|
Tan Liu
|
Tan Liu and Wen-Fan Feng and Zong-Kuan Guo
|
Influence of cosmic expansion on gravitational waveforms
|
19 pages, PRD version
| null |
10.1103/wg63-t8bp
| null |
gr-qc astro-ph.CO
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Gravitational waves undergo redshift as they propagate through the expanding universe, and the redshift may exhibit time-dependent drift. Consequently, for any isolated gravitational wave sources, the mass parameter $\mathcal{M}$ and the redshift $z$ exhibit an observational degeneracy, typically manifesting in the waveform as the redshifted mass $\mathcal{M}(1+z)$. Matching together the wave propagation and the wave generation solutions, we show that dimensionless source parameters depending on mass $\mathcal{M}$ can break this degeneracy. Notably, the postmerger signal from binary neutron stars contains several dimensionless parameters that satisfy this condition, including the quality factors of different frequency components and their frequency ratios. Considering the observations of solely the postmerger signal by the Neutron star Extreme Matter Observatory or the Einstein Telescope, based on the Fisher analysis, we find that the redshift can be measured with fractional uncertainties of $\sim30\%$ for sources at $0.01<z<0.09$. Additionally, we present a corrected derivation of the waveform phase correction due to the redshift drift effect, rectifying a sign error in previous studies.
|
[
{
"version": "v1",
"created": "Fri, 23 May 2025 13:33:20 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 04:54:56 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Liu",
"Tan",
""
],
[
"Feng",
"Wen-Fan",
""
],
[
"Guo",
"Zong-Kuan",
""
]
] |
2505.18252
|
Meredith Durbin
|
Meredith J. Durbin, Yumi Choi, Alessandro Savino, Daniel Weisz, Andrew E. Dolphin, Julianne J. Dalcanton, Myoungwon Jeon, Nitya Kallivayalil, Ting S. Li, Andrew B. Pace, Ekta Patel, Elena Sacchi, Evan D. Skillman, Sangmo Tony Sohn, Roeland P. van der Marel, Andrew Wetzel, Benjamin F. Williams
|
The HST Legacy Archival Uniform Reduction of Local Group Imaging (LAURELIN). I. Photometry and Star Formation Histories for 36 Ultra-faint Dwarf Galaxies
|
25 pages, 13 figures, 4 tables, accepted to ApJ
| null | null | null |
astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
We present uniformly measured resolved stellar photometry and star formation histories (SFHs) for 36 nearby ($\lesssim$ 400 kpc) ultra-faint dwarf galaxies (UFDs; $-7.1 \le M_V \le +0.0$) from new and archival HST imaging. We measure homogeneous distances to all systems via isochrone fitting and find good agreement ($\le$ 2%) for the 18 UFDs that have literature RR Lyrae distances. From the ensemble of SFHs, we find: (i) an average quenching time (here defined as the lookback time by which 80% of the stellar mass formed, $\tau_{80}$) of 12.48 $\pm$ 0.18 Gyr ago ($z = 4.6_{-0.5}^{+0.6}$), which is compatible with reionization-based quenching scenarios; and (ii) modest evidence of a delay ($\lesssim$ 800 Myr) in quenching times of UFDs thought to be satellites of the LMC or on their first infall, relative to long-term Galactic satellites, which is consistent with previous findings. We show that robust SFH measurement via the ancient main sequence turnoff (MSTO) requires a minimum effective luminosity (i.e., luminosity within the observed field of view) of $M_V \leq -2.5$, which corresponds to $\sim$100 stars around the MSTO. We also find that increasing the S/N above $\sim$100 at the MSTO does not improve SFH precision, which remains dominated by stochastic effects associated with the number of available stars. A main challenge driving the precision of UFD SFHs is limitations in the accuracy of foreground dust maps. We make all photometry catalogs public as the first data release of a larger HST archival program targeting all dwarf galaxies within $\sim$1.3 Mpc.
|
[
{
"version": "v1",
"created": "Fri, 23 May 2025 18:00:01 GMT"
},
{
"version": "v2",
"created": "Fri, 29 Aug 2025 18:01:42 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Durbin",
"Meredith J.",
""
],
[
"Choi",
"Yumi",
""
],
[
"Savino",
"Alessandro",
""
],
[
"Weisz",
"Daniel",
""
],
[
"Dolphin",
"Andrew E.",
""
],
[
"Dalcanton",
"Julianne J.",
""
],
[
"Jeon",
"Myoungwon",
""
],
[
"Kallivayalil",
"Nitya",
""
],
[
"Li",
"Ting S.",
""
],
[
"Pace",
"Andrew B.",
""
],
[
"Patel",
"Ekta",
""
],
[
"Sacchi",
"Elena",
""
],
[
"Skillman",
"Evan D.",
""
],
[
"Sohn",
"Sangmo Tony",
""
],
[
"van der Marel",
"Roeland P.",
""
],
[
"Wetzel",
"Andrew",
""
],
[
"Williams",
"Benjamin F.",
""
]
] |
2505.18255
|
Taeho Ryu
|
T. Ryu, R. Sari, S. E. de Mink, O. David, R. Valli, J.-Z. Ma, S. Justham, R. Pakmor, H. Ritter
|
Binary mass transfer in 3D -- Mass Transfer Rate and Morphology
|
17 pages, 12 figures, 2 tables, accepted for publication in A&A, movies available at this Youtube channel: https://www.youtube.com/playlist?list=PLxLK3qI02cQdReUPLcP8HVGmnylV1U-Ou
| null | null | null |
astro-ph.SR astro-ph.HE
|
http://creativecommons.org/licenses/by/4.0/
|
Mass transfer is crucial in binary evolution, yet its theoretical treatment has long relied on analytic models whose key assumptions remain debated. We present a direct and systematic evaluation of these assumptions using high-resolution 3D hydrodynamical simulations including the Coriolis force. We simulate streams overflowing from both the inner and outer Lagrangian points, quantify mass transfer rates, and compare them with analytic solutions. We introduce scaling factors, including the overfilling factor, to render the problem dimensionless. The donor-star models are simplified, with either an isentropic initial stratification and adiabatic evolution or an isothermal structure and evolution, but the scalability of this formulation allows us to extend the results for a mass-transferring system to arbitrarily small overfilling factors for the adiabatic case. We find that the Coriolis force -- often neglected in analytic models -- strongly impacts the stream morphology: breaking axial symmetry, reducing the stream cross section, and shifting its origin toward the donor's trailing side. Contrary to common assumptions, the sonic surface is not flat and does not always intersect the Lagrangian point: instead, it is concave and shifted, particularly toward the accretor's trailing side. Despite these structural asymmetries, mass transfer rates are only mildly suppressed relative to analytic predictions and the deviation is remarkably small -- within a factor of two (ten) for the inner (outer) Lagrangian point over seven orders of magnitude in mass ratio. We use our results to extend the widely-used mass-transfer rate prescriptions by Ritter(1988) and Kolb&Ritter(1990), for both the inner and outer Lagrangian points. These extensions can be readily adopted in stellar evolution codes like MESA, with minimal changes where the original models are already in use.
|
[
{
"version": "v1",
"created": "Fri, 23 May 2025 18:00:03 GMT"
},
{
"version": "v2",
"created": "Wed, 28 May 2025 17:32:11 GMT"
},
{
"version": "v3",
"created": "Sat, 30 Aug 2025 20:31:28 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Ryu",
"T.",
""
],
[
"Sari",
"R.",
""
],
[
"de Mink",
"S. E.",
""
],
[
"David",
"O.",
""
],
[
"Valli",
"R.",
""
],
[
"Ma",
"J. -Z.",
""
],
[
"Justham",
"S.",
""
],
[
"Pakmor",
"R.",
""
],
[
"Ritter",
"H.",
""
]
] |
2505.18336
|
Emiliano Dall'Anese
|
Yiting Chen, Francesco Bullo, Emiliano Dall'Anese
|
Sampled-data Systems: Stability, Contractivity and Single-iteration Suboptimal MPC
|
Modifications relative to version 1: Figure updated
| null | null | null |
eess.SY cs.SY math.OC
|
http://creativecommons.org/licenses/by/4.0/
|
This paper analyzes the stability of interconnected continuous-time (CT) and discrete-time (DT) systems coupled through sampling and zero-order hold mechanisms. The DT system updates its output at regular intervals $T>0$ by applying an $n$-fold composition of a given map. This setup is motivated by online and sampled-data implementations of optimization-based controllers - particularly model predictive control (MPC) - where the DT system models $n$ iterations of an algorithm approximating the solution of an optimization problem.
We introduce the concept of a reduced model, defined as the limiting behavior of the sampled-data system as $T \to 0^+$ and $n \to +\infty$. Our main theoretical contribution establishes that when the reduced model is contractive, there exists a threshold duration $T(n)$ for each iteration count $n$ such that the CT-DT interconnection achieves exponential stability for all sampling periods $T < T(n)$. Finally, under the stronger condition that both the CT and DT systems are contractive, we show exponential stability of their interconnection using a small-gain argument. Our theoretical results provide new insights into suboptimal MPC stability, showing that convergence guarantees hold even when using a single iteration of the optimization algorithm - a practically significant finding for real-time control applications.
|
[
{
"version": "v1",
"created": "Fri, 23 May 2025 19:42:58 GMT"
},
{
"version": "v2",
"created": "Fri, 29 Aug 2025 21:26:58 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Chen",
"Yiting",
""
],
[
"Bullo",
"Francesco",
""
],
[
"Dall'Anese",
"Emiliano",
""
]
] |
2505.18660
|
Zhenglin Huang
|
Zhenglin Huang, Tianxiao Li, Xiangtai Li, Haiquan Wen, Yiwei He, Jiangning Zhang, Hao Fei, Xi Yang, Xiaowei Huang, Bei Peng, Guangliang Cheng
|
So-Fake: Benchmarking and Explaining Social Media Image Forgery Detection
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Recent advances in AI-powered generative models have enabled the creation of increasingly realistic synthetic images, posing significant risks to information integrity and public trust on social media platforms. While robust detection frameworks and diverse, large-scale datasets are essential to mitigate these risks, existing academic efforts remain limited in scope: current datasets lack the diversity, scale, and realism required for social media contexts, while detection methods struggle with generalization to unseen generative technologies. To bridge this gap, we introduce So-Fake-Set, a comprehensive social media-oriented dataset with over 2 million high-quality images, diverse generative sources, and photorealistic imagery synthesized using 35 state-of-the-art generative models. To rigorously evaluate cross-domain robustness, we establish a novel and large-scale (100K) out-of-domain benchmark (So-Fake-OOD) featuring synthetic imagery from commercial models explicitly excluded from the training distribution, creating a realistic testbed for evaluating real-world performance. Leveraging these resources, we present So-Fake-R1, an advanced vision-language framework that employs reinforcement learning for highly accurate forgery detection, precise localization, and explainable inference through interpretable visual rationales. Extensive experiments show that So-Fake-R1 outperforms the second-best method, with a 1.3% gain in detection accuracy and a 4.5% increase in localization IoU. By integrating a scalable dataset, a challenging OOD benchmark, and an advanced detection framework, this work establishes a new foundation for social media-centric forgery detection research. The code, models, and datasets will be released publicly.
|
[
{
"version": "v1",
"created": "Sat, 24 May 2025 11:53:35 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 16:08:09 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Huang",
"Zhenglin",
""
],
[
"Li",
"Tianxiao",
""
],
[
"Li",
"Xiangtai",
""
],
[
"Wen",
"Haiquan",
""
],
[
"He",
"Yiwei",
""
],
[
"Zhang",
"Jiangning",
""
],
[
"Fei",
"Hao",
""
],
[
"Yang",
"Xi",
""
],
[
"Huang",
"Xiaowei",
""
],
[
"Peng",
"Bei",
""
],
[
"Cheng",
"Guangliang",
""
]
] |
2505.18665
|
Aaron Beyen
|
Aaron Beyen, Christian Maes and Ji-Hui Pei
|
Coupling an elastic string to an active bath: the emergence of inverse damping
|
5 figures
| null | null | null |
cond-mat.stat-mech
|
http://creativecommons.org/licenses/by/4.0/
|
We consider a slow elastic string with Klein-Gordon dynamics coupled to a bath of run-and-tumble particles. We derive and solve the induced Langevin-Klein-Gordon string dynamics with explicit expressions for the streaming term, friction coefficient, and noise variance. These parameters are computed exactly in a weak coupling expansion. The induced friction is a sum of two terms: one entropic, proportional to the noise variance as in the Einstein relation for a thermal equilibrium bath, and a frenetic contribution that can take both signs. The frenetic part wins for higher bath persistence, making the total friction negative, and hence creating a wave instability akin to inverse Landau damping. However, this acceleration decreases and eventually disappears when the propulsion speed of the active particles becomes much higher. Detailed simulations confirm the initial growth driven by this anti-damping.
|
[
{
"version": "v1",
"created": "Sat, 24 May 2025 11:59:44 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 21:53:00 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Beyen",
"Aaron",
""
],
[
"Maes",
"Christian",
""
],
[
"Pei",
"Ji-Hui",
""
]
] |
2505.18906
|
Abhijit Chakraborty
|
Abhijit Chakraborty and Chahana Dahal and Vivek Gupta
|
Federated Retrieval-Augmented Generation: A Systematic Mapping Study
| null | null | null | null |
cs.CL cs.IR
|
http://creativecommons.org/licenses/by/4.0/
|
Federated Retrieval-Augmented Generation (Federated RAG) combines Federated Learning (FL), which enables distributed model training without exposing raw data, with Retrieval-Augmented Generation (RAG), which improves the factual accuracy of language models by grounding outputs in external knowledge. As large language models are increasingly deployed in privacy-sensitive domains such as healthcare, finance, and personalized assistance, Federated RAG offers a promising framework for secure, knowledge-intensive natural language processing (NLP). To the best of our knowledge, this paper presents the first systematic mapping study of Federated RAG, covering literature published between 2020 and 2025. Following Kitchenham's guidelines for evidence-based software engineering, we develop a structured classification of research focuses, contribution types, and application domains. We analyze architectural patterns, temporal trends, and key challenges, including privacy-preserving retrieval, cross-client heterogeneity, and evaluation limitations. Our findings synthesize a rapidly evolving body of research, identify recurring design patterns, and surface open questions, providing a foundation for future work at the intersection of RAG and federated systems.
|
[
{
"version": "v1",
"created": "Sat, 24 May 2025 23:45:12 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 06:08:11 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Chakraborty",
"Abhijit",
""
],
[
"Dahal",
"Chahana",
""
],
[
"Gupta",
"Vivek",
""
]
] |
2505.19030
|
Zhengkang Guo
|
Zhengkang Guo, Wenhao Liu, Mingchen Xie, Jingwen Xu, Zisu Huang, Muzhao Tian, Jianhan Xu, Muling Wu, Xiaohua Wang, Changze Lv, He-Da Wang, Hu Yao, Xiaoqing Zheng, Xuanjing Huang
|
RECAST: Strengthening LLMs' Complex Instruction Following with Constraint-Verifiable Data
| null | null | null | null |
cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Large language models (LLMs) are increasingly expected to tackle complex tasks, driven by their expanding applications and users' growing proficiency in crafting sophisticated prompts. However, as the number of explicitly stated requirements increases (particularly more than 10 constraints), LLMs often struggle to accurately follow such complex instructions. To address this challenge, we propose RECAST, a novel framework for synthesizing datasets where each example incorporates far more constraints than those in existing benchmarks. These constraints are extracted from real-world prompt-response pairs to ensure practical relevance. RECAST enables automatic verification of constraint satisfaction via rule-based validators for quantitative constraints and LLM-based validators for qualitative ones. Using this framework, we construct RECAST-30K, a large-scale, high-quality dataset comprising 30k instances spanning 15 constraint types. Experimental results demonstrate that models fine-tuned on RECAST-30K show substantial improvements in following complex instructions. Moreover, the verifiability provided by RECAST enables the design of reward functions for reinforcement learning, which further boosts model performance on complex and challenging tasks.
|
[
{
"version": "v1",
"created": "Sun, 25 May 2025 08:31:08 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 07:53:33 GMT"
},
{
"version": "v3",
"created": "Sun, 31 Aug 2025 08:01:12 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Guo",
"Zhengkang",
""
],
[
"Liu",
"Wenhao",
""
],
[
"Xie",
"Mingchen",
""
],
[
"Xu",
"Jingwen",
""
],
[
"Huang",
"Zisu",
""
],
[
"Tian",
"Muzhao",
""
],
[
"Xu",
"Jianhan",
""
],
[
"Wu",
"Muling",
""
],
[
"Wang",
"Xiaohua",
""
],
[
"Lv",
"Changze",
""
],
[
"Wang",
"He-Da",
""
],
[
"Yao",
"Hu",
""
],
[
"Zheng",
"Xiaoqing",
""
],
[
"Huang",
"Xuanjing",
""
]
] |
2505.19412
|
Nagare Katayama
|
Nagare Katayama, Yuya Tanizaki
|
2d Cardy-Rabinovici model with the modified Villain lattice: Exact dualities and symmetries
|
30 pages, 4 figures
| null | null |
YITP-25-78
|
hep-th hep-lat
|
http://creativecommons.org/licenses/by/4.0/
|
The Cardy-Rabinovici model is a toy model of the lattice $U(1)$ gauge theories to study various oblique confinement states associated with the nonzero $\theta$ angles. We reformulate the $2$d version of this model using the modified Villain lattice formalism, and we establish the exact $\theta$ periodicity for the Witten effect and the strong-weak duality at the finite lattice spacings. We then study the phase structure of this model based on the duality, symmetry and anomaly, and the perturbative renormalization group.
|
[
{
"version": "v1",
"created": "Mon, 26 May 2025 02:04:31 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 09:04:46 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 07:11:43 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Katayama",
"Nagare",
""
],
[
"Tanizaki",
"Yuya",
""
]
] |
2505.19691
|
Chon Man Sou
|
Chon Man Sou
|
Alternative Derivations of Hawking Radiation
|
40 pages, 3 figures. Contribution to a chapter of the book "The Black Hole Information Paradox", edited by A. Akil and C. Bambi
| null |
10.1007/978-981-96-6170-1_2
| null |
gr-qc
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Since the original derivation of Hawking radiation, there have been lots of alternative approaches to show the same fact that black holes emit particles as hot bodies with a temperature. These alternative methods generally rely on different conditions and physical quantities to manifest the radiation, providing various points of view of this effect in the intersection of gravity and quantum theory. This chapter presents some alternative derivations of Hawking radiation in the literature, including the tunneling, anomaly and Green's function methods. From these methods, various features of the black hole system can be seen, such as the gravitational and trace anomalies of the $(1+1)$-dimensional effective theory and the analytical continuation of the complexified spacetime.
|
[
{
"version": "v1",
"created": "Mon, 26 May 2025 08:49:56 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Sou",
"Chon Man",
""
]
] |
2505.19777
|
Tao Yu
|
Tao Yu, Yunsong Ning, Yi Yuan, Shihan Zhao, Songran Qi, Minchen Sun, Yuye Li, Zhirui Liu, Aiyu Bai, Hesheng Liu, Yibo Lin, Geng Tuo, Ting On Chan, Zhou Zhou, Yu Chen, Yu Chen, Jian Tang
|
MuGrid-v2: A novel scintillator detector for multidisciplinary applications
| null | null |
10.1063/5.0273373
| null |
physics.ins-det hep-ex
|
http://creativecommons.org/licenses/by/4.0/
|
Muography, traditionally recognized as a potent instrument for imaging the internal structure of gigantic objects, has initialized various interdisciplinary applications. As the financial and labor costs of muography detector development hinder their massive applications, we develop a novel muon detector called MuGrid by coupling a monolithic plastic scintillator with the light guide array in order to achieve competitive spatial resolution while substantially reducing production costs. For a prototype detector in 30 cm $\times$ 30 cm, the intrinsic spatial resolution has been optimized toward a millimeter scale. An outdoor field muography experiment was conducted to monitor two buildings for validation purposes. The test successfully resolved the geometric influence of architectural features based on the attenuation of muon flux in a good agreement between experimental results and the simulation prediction.
|
[
{
"version": "v1",
"created": "Mon, 26 May 2025 10:01:55 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Yu",
"Tao",
""
],
[
"Ning",
"Yunsong",
""
],
[
"Yuan",
"Yi",
""
],
[
"Zhao",
"Shihan",
""
],
[
"Qi",
"Songran",
""
],
[
"Sun",
"Minchen",
""
],
[
"Li",
"Yuye",
""
],
[
"Liu",
"Zhirui",
""
],
[
"Bai",
"Aiyu",
""
],
[
"Liu",
"Hesheng",
""
],
[
"Lin",
"Yibo",
""
],
[
"Tuo",
"Geng",
""
],
[
"Chan",
"Ting On",
""
],
[
"Zhou",
"Zhou",
""
],
[
"Chen",
"Yu",
""
],
[
"Chen",
"Yu",
""
],
[
"Tang",
"Jian",
""
]
] |
2505.19904
|
Zsolt Szab\'o
|
Zsolt Szab\'o, Stefan Gehr, Paolo Facchi, Kazuya Yuasa, Daniel Burgarth, Davide Lonigro
|
Robust quantification of spectral transitions in perturbed quantum systems
|
19 pages, 1 figure
|
Phys. Rev. A 112 (2025), 032202
|
10.1103/j44p-13j7
| null |
quant-ph math-ph math.MP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
A quantum system subject to an external perturbation can experience leakage between uncoupled regions of its energy spectrum separated by a gap. To quantify this phenomenon, we present two complementary results. First, we establish time-independent bounds on the distances between the true dynamics and the dynamics generated by block-diagonal effective evolutions constructed via the Schrieffer-Wolff and Bloch methods. Second, we prove that, under the right conditions, this leakage remains small eternally. That is, we derive a time-independent bound on the leakage itself, expressed in terms of the spectral gap of the unperturbed Hamiltonian and the norm of the perturbation, ensuring its validity for arbitrarily large times. Our approach only requires a finite spectral gap, thus accommodating continuous and unbounded spectra. Finally, we apply our bounds to specific systems of practical interest.
|
[
{
"version": "v1",
"created": "Mon, 26 May 2025 12:33:07 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Szabó",
"Zsolt",
""
],
[
"Gehr",
"Stefan",
""
],
[
"Facchi",
"Paolo",
""
],
[
"Yuasa",
"Kazuya",
""
],
[
"Burgarth",
"Daniel",
""
],
[
"Lonigro",
"Davide",
""
]
] |
2505.20404
|
Sha Yi
|
Sha Yi, Xueqian Bai, Adabhav Singh, Jianglong Ye, Michael T Tolley, Xiaolong Wang
|
Co-Design of Soft Gripper with Neural Physics
| null | null | null | null |
cs.RO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
For robot manipulation, both the controller and end-effector design are crucial. Soft grippers are generalizable by deforming to different geometries, but designing such a gripper and finding its grasp pose remains challenging. In this paper, we propose a co-design framework that generates an optimized soft gripper's block-wise stiffness distribution and its grasping pose, using a neural physics model trained in simulation. We derived a uniform-pressure tendon model for a flexure-based soft finger, then generated a diverse dataset by randomizing both gripper pose and design parameters. A neural network is trained to approximate this forward simulation, yielding a fast, differentiable surrogate. We embed that surrogate in an end-to-end optimization loop to optimize the ideal stiffness configuration and best grasp pose. Finally, we 3D-print the optimized grippers of various stiffness by changing the structural parameters. We demonstrate that our co-designed grippers significantly outperform baseline designs in both simulation and hardware experiments. More info: http://yswhynot.github.io/codesign-soft/
|
[
{
"version": "v1",
"created": "Mon, 26 May 2025 18:00:09 GMT"
},
{
"version": "v2",
"created": "Fri, 30 May 2025 01:41:00 GMT"
},
{
"version": "v3",
"created": "Tue, 02 Sep 2025 05:43:33 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Yi",
"Sha",
""
],
[
"Bai",
"Xueqian",
""
],
[
"Singh",
"Adabhav",
""
],
[
"Ye",
"Jianglong",
""
],
[
"Tolley",
"Michael T",
""
],
[
"Wang",
"Xiaolong",
""
]
] |
2505.20579
|
Dane Malenfant
|
Dane Malenfant, Blake A. Richards
|
The challenge of hidden gifts in multi-agent reinforcement learning
|
Updated proof section and included single agent baseline for key-to-door in the appendix. Related work now is part of the main section
| null | null | null |
cs.LG cs.AI cs.MA
|
http://creativecommons.org/licenses/by/4.0/
|
Sometimes we benefit from actions that others have taken even when we are unaware that they took those actions. For example, if your neighbor chooses not to take a parking spot in front of your house when you are not there, you can benefit, even without being aware that they took this action. These "hidden gifts" represent an interesting challenge for multi-agent reinforcement learning (MARL), since assigning credit when the beneficial actions of others are hidden is non-trivial. Here, we study the impact of hidden gifts with a very simple MARL task. In this task, agents in a grid-world environment have individual doors to unlock in order to obtain individual rewards. As well, if all the agents unlock their door the group receives a larger collective reward. However, there is only one key for all of the doors, such that the collective reward can only be obtained when the agents drop the key for others after they use it. Notably, there is nothing to indicate to an agent that the other agents have dropped the key, thus the act of dropping the key for others is a "hidden gift". We show that several different state-of-the-art RL algorithms, including MARL algorithms, fail to learn how to obtain the collective reward in this simple task. Interestingly, we find that independent model-free policy gradient agents can solve the task when we provide them with information about their own action history, but MARL agents still cannot solve the task with action history. Finally, we derive a correction term for these independent agents, inspired by learning aware approaches, which reduces the variance in learning and helps them to converge to collective success more reliably. These results show that credit assignment in multi-agent settings can be particularly challenging in the presence of "hidden gifts", and demonstrate that learning awareness in independent agents can benefit these settings.
|
[
{
"version": "v1",
"created": "Mon, 26 May 2025 23:28:52 GMT"
},
{
"version": "v2",
"created": "Thu, 29 May 2025 13:37:25 GMT"
},
{
"version": "v3",
"created": "Wed, 30 Jul 2025 01:18:05 GMT"
},
{
"version": "v4",
"created": "Mon, 01 Sep 2025 16:56:59 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Malenfant",
"Dane",
""
],
[
"Richards",
"Blake A.",
""
]
] |
2505.20988
|
Andr\'es La\'in Sanclemente
|
Diego C\'ordoba, Andr\'es La\'in-Sanclemente, Luis Mart\'inez-Zoroa
|
Finite-time singularity via multi-layer degenerate pendula for the 2D Boussinesq equation with uniform $C^{1,\sqrt{\frac{4}{3}}-1-\epsilon}\cap L^2$ force
|
Some errata corrected
|
Advances in Mathematics, Volume 480, Part A, 2025, 110480, ISSN 0001-8708
|
10.1016/j.aim.2025.110480
| null |
math.AP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We establish the existence of compactly supported solutions of the inviscid incompressible 2D Boussinesq equation with $C^{1,\sqrt{\frac{4}{3}}-1-\varepsilon}\cap L^{2}$ force that develop a singularity in finite time. Importantly, the force preserves this regularity at the blow-up time. Moreover, the forces in the vorticity and density equations have compact support. The mechanism behind the blow-up is an accumulated hysteresis effect on the vorticity caused by an infinite chain of "degenerate" pendula and flickering density.
|
[
{
"version": "v1",
"created": "Tue, 27 May 2025 10:23:44 GMT"
},
{
"version": "v2",
"created": "Fri, 04 Jul 2025 16:03:36 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 09:01:11 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Córdoba",
"Diego",
""
],
[
"Laín-Sanclemente",
"Andrés",
""
],
[
"Martínez-Zoroa",
"Luis",
""
]
] |
2505.21541
|
Hang Zhao
|
Zitong Wang, Hang Zhao, Qianyu Zhou, Xuequan Lu, Xiangtai Li, Yiren Song
|
DiffDecompose: Layer-Wise Decomposition of Alpha-Composited Images via Diffusion Transformers
| null | null | null | null |
cs.CV cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Diffusion models have recently motivated great success in many generation tasks like object removal. Nevertheless, existing image decomposition methods struggle to disentangle semi-transparent or transparent layer occlusions due to mask prior dependencies, static object assumptions, and the lack of datasets. In this paper, we delve into a novel task: Layer-Wise Decomposition of Alpha-Composited Images, aiming to recover constituent layers from single overlapped images under the condition of semi-transparent/transparent alpha layer non-linear occlusion. To address challenges in layer ambiguity, generalization, and data scarcity, we first introduce AlphaBlend, the first large-scale and high-quality dataset for transparent and semi-transparent layer decomposition, supporting six real-world subtasks (e.g., translucent flare removal, semi-transparent cell decomposition, glassware decomposition). Building on this dataset, we present DiffDecompose, a diffusion Transformer-based framework that learns the posterior over possible layer decompositions conditioned on the input image, semantic prompts, and blending type. Rather than regressing alpha mattes directly, DiffDecompose performs In-Context Decomposition, enabling the model to predict one or multiple layers without per-layer supervision, and introduces Layer Position Encoding Cloning to maintain pixel-level correspondence across layers. Extensive experiments on the proposed AlphaBlend dataset and public LOGO dataset verify the effectiveness of DiffDecompose. The code and dataset will be available upon paper acceptance. Our code will be available at: https://github.com/Wangzt1121/DiffDecompose.
|
[
{
"version": "v1",
"created": "Sat, 24 May 2025 16:08:04 GMT"
},
{
"version": "v2",
"created": "Fri, 30 May 2025 13:38:16 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 09:14:35 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Wang",
"Zitong",
""
],
[
"Zhao",
"Hang",
""
],
[
"Zhou",
"Qianyu",
""
],
[
"Lu",
"Xuequan",
""
],
[
"Li",
"Xiangtai",
""
],
[
"Song",
"Yiren",
""
]
] |
2505.21612
|
Rossella Gamba
|
Rossella Gamba, Jacob Lange, Danilo Chiaramello, Jacopo Tissino and Snehal Tibrewal
|
Revisiting GW150914 with a non-planar, eccentric waveform model
|
13 pages, 6 figures; version accepted for publication
| null | null | null |
gr-qc astro-ph.HE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The first direct detection of gravitational waves by the LIGO collaboration, GW150914, marked the start of a new exciting era in astronomy, enabling the study of the Universe through a new messenger. Since then, the field has grown rapidly, with the development of increasingly more sophisticated techniques to detect, analyze and interpret the signals. In this paper we revisit GW150914, presenting updated estimates of its source parameters using a waveform model developed within the EOB formalism, able to describe gravitational-wave emission from generic non-circular, non-planar binaries. We provide a comprehensive analysis of the signal and its properties, considering and contrasting various scenarios for the source: from the simplest, aligned-spin quasi-circular binary black hole merger, to more complex scenarios, including precession, eccentricity or both. Unsurprisingly, we find that the signal is consistent with a quasi-circular ($e < 0.08$ at $15$ Hz), slowly spinning $(\chi_{\rm eff} = -0.03^{+0.12}_{-0.13})$ binary black hole merger, a-posteriori validating a considerable body of works. This is the first analysis performed with an inspiral-merger-ringdown model containing both eccentricity and precession.
|
[
{
"version": "v1",
"created": "Tue, 27 May 2025 18:00:00 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 08:12:10 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Gamba",
"Rossella",
""
],
[
"Lange",
"Jacob",
""
],
[
"Chiaramello",
"Danilo",
""
],
[
"Tissino",
"Jacopo",
""
],
[
"Tibrewal",
"Snehal",
""
]
] |
2505.21660
|
Xiaojie Xu
|
Xiaojie Xu, Xinli Xu, Sirui Chen, Haoyu Chen, Fan Zhang, Ying-Cong Chen
|
PreGenie: An Agentic Framework for High-quality Visual Presentation Generation
|
Accepted at EMNLP 2025, Findings
| null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Visual presentations are vital for effective communication. Early attempts to automate their creation using deep learning often faced issues such as poorly organized layouts, inaccurate text summarization, and a lack of image understanding, leading to mismatched visuals and text. These limitations restrict their application in formal contexts like business and scientific research. To address these challenges, we propose PreGenie, an agentic and modular framework powered by multimodal large language models (MLLMs) for generating high-quality visual presentations.
PreGenie is built on the Slidev presentation framework, where slides are rendered from Markdown code. It operates in two stages: (1) Analysis and Initial Generation, which summarizes multimodal input and generates initial code, and (2) Review and Re-generation, which iteratively reviews intermediate code and rendered slides to produce final, high-quality presentations. Each stage leverages multiple MLLMs that collaborate and share information. Comprehensive experiments demonstrate that PreGenie excels in multimodal understanding, outperforming existing models in both aesthetics and content consistency, while aligning more closely with human design preferences.
|
[
{
"version": "v1",
"created": "Tue, 27 May 2025 18:36:19 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 00:34:05 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Xu",
"Xiaojie",
""
],
[
"Xu",
"Xinli",
""
],
[
"Chen",
"Sirui",
""
],
[
"Chen",
"Haoyu",
""
],
[
"Zhang",
"Fan",
""
],
[
"Chen",
"Ying-Cong",
""
]
] |
2505.21687
|
Jiemin Li
|
Jiemin Li, Yanhong Gu, Takemi Yamada, Zebin Wu, Genda Gu, Tonica Valla, Ilya Drozdov, Ivan Bozovic, Mark P. M. Dean, Takami Tohyama, Jonathan Pelliciari and Valentina Bisogni
|
Anisotropic electron damping and energy gap in Bi$_2$Sr$_2$CaCu$_2$O$_{8+\delta}$
| null |
Phys. Rev. Research 7, 023319 (2025)
|
10.1103/k32g-pknp
| null |
cond-mat.str-el
|
http://creativecommons.org/licenses/by/4.0/
|
The many body electron-electron interaction in cuprates causes the broadening of the electronic bands in \textit{\textbf{k}}-space, leading to a deviation from the standard Fermi liquid. While a \textit{\textbf{k}}-dependent anisotropic electronic scattering (\textit{\textbf{k}}-DAES) has been assessed by photoemission, its fingerprint in \textit{\textbf{Q}}-space has been scarcely considered. Here, we explore the \textit{\textbf{Q}}-dependent electron dynamics in optimally doped Bi$_2$Sr$_2$CaCu$_2$O$_{8+\delta}$ through the evolution of low-energy charge excitations as measured by resonant inelastic x ray scattering (RIXS). In the normal state, the RIXS spectra display a continuum of excitations down to 0~meV, while the superconducting state features a spectral weight suppression below 80 meV without any enhancement at higher energies. To interpret the energy and \textit{\textbf{Q}}-evolution of our data, we introduce a phenomenological expression of the charge susceptibility by including the \textit{\textbf{k}}-DAES. We show that only the charge susceptibility with \textit{\textbf{k}}-DAES captures the RIXS data, highlighting the importance of \textit{\textbf{k}}-DAES when describing the \textit{\textbf{Q}}-dependence of charge excitations from 0 to few eV scale. Furthermore, we also find that the inclusion of \textit{\textbf{k}}-DAES is essential when quantitative parameters such as the electronic energy gap are extracted from RIXS data.
|
[
{
"version": "v1",
"created": "Tue, 27 May 2025 19:20:43 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Li",
"Jiemin",
""
],
[
"Gu",
"Yanhong",
""
],
[
"Yamada",
"Takemi",
""
],
[
"Wu",
"Zebin",
""
],
[
"Gu",
"Genda",
""
],
[
"Valla",
"Tonica",
""
],
[
"Drozdov",
"Ilya",
""
],
[
"Bozovic",
"Ivan",
""
],
[
"Dean",
"Mark P. M.",
""
],
[
"Tohyama",
"Takami",
""
],
[
"Pelliciari",
"Jonathan",
""
],
[
"Bisogni",
"Valentina",
""
]
] |
2505.22166
|
Takehisa Hasegawa
|
Kazuki Wataya, Takehisa Hasegawa
|
Critical and Nonpercolating Phases in Bond Percolation on the Song-Havlin-Makse Network
|
12 pages, 8 figures
| null | null | null |
cond-mat.stat-mech cond-mat.dis-nn
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We investigate bond percolation on the Song-Havlin-Makse (SHM) network, a scale-free tree with a tunable degree exponent and dimensionality. Using a generating function approach, we analytically derive the average size and the fractal exponent of the root cluster for deterministic cases. Our analysis reveals that bond percolation on the SHM network remains in a nonpercolating phase for all $p < 1$ when the network is fractal (i.e., finite-dimensional), whereas it exhibits a critical phase, where the cluster size distribution follows a power-law with a $p$-dependent exponent, throughout the entire range of $p$ when the network is small-world (i.e., infinite-dimensional), regardless of the specific dimensionality or degree exponent. The analytical results are in excellent agreement with Monte Carlo simulations.
|
[
{
"version": "v1",
"created": "Wed, 28 May 2025 09:33:11 GMT"
},
{
"version": "v2",
"created": "Thu, 29 May 2025 06:12:33 GMT"
},
{
"version": "v3",
"created": "Tue, 02 Sep 2025 05:51:39 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Wataya",
"Kazuki",
""
],
[
"Hasegawa",
"Takehisa",
""
]
] |
2505.22186
|
Mark Fortune
|
Mark Fortune, Neale P. Gibson, Hannah Diamond-Lowe, Jo\~ao M. Mendon\c{c}a, Am\'elie Gressier, Daniel Kitzmann, Natalie H. Allen, Prune C. August, Jegug Ih, Erik Meier Vald\'es, Merlin Zgraggen, Lars A. Buchhave, Brice-Olivier Demory, N\'estor Espinoza, Kevin Heng, Kathryn Jones, Alexander D. Rathcke
|
Hot Rocks Survey III: A deep eclipse for LHS 1140c and a new Gaussian process method to account for correlated noise in individual pixels
|
Accepted for publication in Astronomy & Astrophysics (updated version after language edits). 35 pages, 28 figures
|
A&A 701, A25 (2025)
|
10.1051/0004-6361/202554198
| null |
astro-ph.EP astro-ph.IM
|
http://creativecommons.org/licenses/by/4.0/
|
Time-series photometry at mid-infrared wavelengths is becoming a common technique to search for atmospheres around rocky exoplanets. This method constrains the brightness temperature of the planet to determine whether heat redistribution is taking place - indicative of an atmosphere - or whether the heat is reradiated from a low albedo bare rock. By observing at 15$\mu$m we are also highly sensitive to CO$_2$ absorption. We observed three eclipses of the rocky super-Earth LHS 1140c using MIRI/Imaging with the F1500W filter. We found significant variation in the initial settling ramp for these observations and identify a potential trend between detector settling and the previous filter used by MIRI. We analysed our data using aperture photometry but also developed a novel approach which joint-fits pixel light curves directly using a shared eclipse model and a flexible multi-dimensional Gaussian process which models changes in the PSF over time. We demonstrate using simulated data that our method has the ability to weight away from particular pixels which show increased systematics, allowing for the recovery of eclipse depths in a more robust and precise way. Both methods and an independent analysis detect the eclipse at $>5\sigma$ and are highly consistent with a low albedo bare rock. We recover a dayside brightness temperature of $T_\mathrm{day} = 561\pm44$ K, close to the theoretical maximum of $T_\text{day; max} = 537\pm9$ K. We rule out a wide range of atmospheric forward models to $>3\sigma$ including pure CO$_2$ atmospheres with surface pressure $\ge10$ mbar and pure H$_2$O atmospheres with surface pressure $\ge1$ bar. Our strict constraints on potential atmospheric composition, in combination with future observations of the exciting outer planet LHS 1140b, could provide a powerful benchmark to understand atmospheric escape around M dwarfs.
|
[
{
"version": "v1",
"created": "Wed, 28 May 2025 09:59:03 GMT"
},
{
"version": "v2",
"created": "Thu, 31 Jul 2025 10:43:53 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Fortune",
"Mark",
""
],
[
"Gibson",
"Neale P.",
""
],
[
"Diamond-Lowe",
"Hannah",
""
],
[
"Mendonça",
"João M.",
""
],
[
"Gressier",
"Amélie",
""
],
[
"Kitzmann",
"Daniel",
""
],
[
"Allen",
"Natalie H.",
""
],
[
"August",
"Prune C.",
""
],
[
"Ih",
"Jegug",
""
],
[
"Valdés",
"Erik Meier",
""
],
[
"Zgraggen",
"Merlin",
""
],
[
"Buchhave",
"Lars A.",
""
],
[
"Demory",
"Brice-Olivier",
""
],
[
"Espinoza",
"Néstor",
""
],
[
"Heng",
"Kevin",
""
],
[
"Jones",
"Kathryn",
""
],
[
"Rathcke",
"Alexander D.",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.