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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2502.18000
|
Yuguang Shi
|
Shihang He, Yuguang Shi, Haobin Yu
|
Positive mass theorems on singular spaces and some applications
|
62 pages, 7 figures, all comments are welcome!
| null | null | null |
math.DG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Building upon dimension reduction techniques in the study of positive scalar curvature (PSC) geometry, we prove an effective version of the positive mass theorem (PMT) for asymptotically flat (AF) manifolds of dimension $n\leq 8$ with arbitrary ends (Theorem \ref{thm: 8dim Schoen conj}). Furthermore, we prove two "free of singularity type rigidity theorems" for minimal hypersurfaces with isolated singularities (Theorem \ref{prop: rigidity for minimal surface} and Theorem \ref{thm: georch free of singularity}). Our approach bypasses the need for N. Smale's regularity theorem for minimal hypersurfaces in generic $8$-dimensional compact manifolds, providing a direct derivation of the PMT for such AF manifolds (Theorem \ref{thm: pmt8dim}).
Motivated by these developments, we further establish PMT for singular spaces (Theorems \ref{thm:pmt with singularity4}). These results assume only that the scalar curvature is non-negative in a strong spectral sense, a condition naturally aligned with the stability of minimal hypersurfaces in PSC ambient manifolds.
|
[
{
"version": "v1",
"created": "Tue, 25 Feb 2025 09:08:29 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 08:27:52 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"He",
"Shihang",
""
],
[
"Shi",
"Yuguang",
""
],
[
"Yu",
"Haobin",
""
]
] |
2502.18289
|
Natalia Bondarenko
|
Natalia P. Bondarenko
|
Uniform stability for the inverse Sturm-Liouville problem with eigenparameter-dependent boundary conditions
| null |
J. Math. Phys. 66 (2025), 063503
|
10.1063/5.0267682
| null |
math.SP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We consider a class of self-adjoint Sturm-Liouville problems with rational functions of the spectral parameter in the boundary conditions. The uniform stability for direct and inverse spectral problems is proved for the first time for Sturm-Liouville operator pencils with boundary conditions depending on the eigenparameter. Furthermore, we obtain stability estimates for finite data approximations, which are important from the practical viewpoint. Our method is based on Darboux-type transforms and proving of their Lipschitz continuity.
|
[
{
"version": "v1",
"created": "Tue, 25 Feb 2025 15:27:44 GMT"
},
{
"version": "v2",
"created": "Thu, 10 Apr 2025 19:29:55 GMT"
},
{
"version": "v3",
"created": "Sun, 31 Aug 2025 12:47:07 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Bondarenko",
"Natalia P.",
""
]
] |
2502.18517
|
Jianwei Wang
|
Jianwei Wang, Chengming Shi, Junyao Yang, Haoran Li, Qianli Ma, Huiping Zhuang, Cen Chen and Ziqian Zeng
|
RewardDS: Privacy-Preserving Fine-Tuning for Large Language Models via Reward Driven Data Synthesis
|
Accepted by EMNLP2025 Main
| null | null | null |
cs.CR cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
The success of large language models (LLMs) has attracted many individuals to fine-tune them for domain-specific tasks by uploading their data. However, in sensitive areas like healthcare and finance, privacy concerns often arise. One promising solution is to generate synthetic data with Differential Privacy (DP) guarantees to replace private data. However, these synthetic data contain significant flawed data, which are considered as noise. Existing solutions typically rely on naive filtering by comparing ROUGE-L scores or embedding similarities, which are ineffective in addressing the noise. To address this issue, we propose \textit{RewardDS}, a novel privacy-preserving framework that fine-tunes a reward proxy model and uses reward signals to guide the synthetic data generation. Our \textit{RewardDS} introduces two key modules, Reward Guided Filtering and Self-Optimizing Refinement, to both filter and refine the synthetic data, effectively mitigating the noise. Extensive experiments across medical, financial, and code generation domains demonstrate the effectiveness of our method.
|
[
{
"version": "v1",
"created": "Sun, 23 Feb 2025 02:52:23 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 08:34:15 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Wang",
"Jianwei",
""
],
[
"Shi",
"Chengming",
""
],
[
"Yang",
"Junyao",
""
],
[
"Li",
"Haoran",
""
],
[
"Ma",
"Qianli",
""
],
[
"Zhuang",
"Huiping",
""
],
[
"Chen",
"Cen",
""
],
[
"Zeng",
"Ziqian",
""
]
] |
2502.18601
|
Teddy Lazebnik Prof.
|
Uri Itai, Asael Bar Ilan, Teddy Lazebnik
|
Tighten The Lasso: A Convex Hull Volume-based Anomaly Detection Method
| null | null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Detecting out-of-distribution (OOD) data is a critical task for maintaining model reliability and robustness. In this study, we propose a novel anomaly detection algorithm that leverages the convex hull (CH) property of a dataset by exploiting the observation that OOD samples marginally increase the CH's volume compared to in-distribution samples. Thus, we establish a decision boundary between OOD and in-distribution data by iteratively computing the CH's volume as samples are removed, stopping when such removal does not significantly alter the CH's volume. The proposed algorithm is evaluated against seven widely used anomaly detection methods across ten datasets, demonstrating performance comparable to state-of-the-art (SOTA) techniques. Furthermore, we introduce a computationally efficient criterion for identifying datasets where the proposed method outperforms existing SOTA approaches.
|
[
{
"version": "v1",
"created": "Tue, 25 Feb 2025 19:39:20 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 11:50:29 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Itai",
"Uri",
""
],
[
"Ilan",
"Asael Bar",
""
],
[
"Lazebnik",
"Teddy",
""
]
] |
2502.18686
|
Antti Kykk\"anen
|
Joonas Ilmavirta, Antti Kykk\"anen, Teemu Saksala
|
The elastic ray transform
|
23 pages, updated introduction
| null | null | null |
math.FA math.AP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We introduce and study a new family of tensor tomography problems. At rank 2 it corresponds to linearization of travel time of elastic waves, measured for all polarizations. We provide a kernel characterization for ranks up to 2. The kernels consist of potential tensors, but in an unusual sense: the associated differential operators have degree 2 instead of the familiar 1. The proofs are based on Fourier analysis, Helmholtz decompositions, and cohomology.
|
[
{
"version": "v1",
"created": "Tue, 25 Feb 2025 22:49:50 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 17:40:55 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Ilmavirta",
"Joonas",
""
],
[
"Kykkänen",
"Antti",
""
],
[
"Saksala",
"Teemu",
""
]
] |
2502.18826
|
Yuxiao Wen
|
Yuxiao Wen
|
Adversarial Combinatorial Semi-bandits with Graph Feedback
|
To appear in ICML 2025
| null | null | null |
cs.LG cs.IT math.IT stat.ML
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In combinatorial semi-bandits, a learner repeatedly selects from a combinatorial decision set of arms, receives the realized sum of rewards, and observes the rewards of the individual selected arms as feedback. In this paper, we extend this framework to include \emph{graph feedback}, where the learner observes the rewards of all neighboring arms of the selected arms in a feedback graph $G$. We establish that the optimal regret over a time horizon $T$ scales as $\widetilde{\Theta}(S\sqrt{T}+\sqrt{\alpha ST})$, where $S$ is the size of the combinatorial decisions and $\alpha$ is the independence number of $G$. This result interpolates between the known regrets $\widetilde\Theta(S\sqrt{T})$ under full information (i.e., $G$ is complete) and $\widetilde\Theta(\sqrt{KST})$ under the semi-bandit feedback (i.e., $G$ has only self-loops), where $K$ is the total number of arms. A key technical ingredient is to realize a convexified action using a random decision vector with negative correlations. We also show that online stochastic mirror descent (OSMD) that only realizes convexified actions in expectation is suboptimal. In addition, we describe the problem of \emph{combinatorial semi-bandits with general capacity} and apply our results to derive an improved regret upper bound, which may be of independent interest.
|
[
{
"version": "v1",
"created": "Wed, 26 Feb 2025 05:01:11 GMT"
},
{
"version": "v2",
"created": "Fri, 28 Feb 2025 00:20:14 GMT"
},
{
"version": "v3",
"created": "Fri, 02 May 2025 03:02:10 GMT"
},
{
"version": "v4",
"created": "Thu, 05 Jun 2025 05:09:33 GMT"
},
{
"version": "v5",
"created": "Sat, 26 Jul 2025 04:49:04 GMT"
},
{
"version": "v6",
"created": "Mon, 01 Sep 2025 16:16:01 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Wen",
"Yuxiao",
""
]
] |
2502.18848
|
Kerem Zaman
|
Kerem Zaman and Shashank Srivastava
|
A Causal Lens for Evaluating Faithfulness Metrics
|
25 pages, 22 figures, 9 tables
| null | null | null |
cs.CL cs.AI cs.LG stat.ME
|
http://creativecommons.org/licenses/by/4.0/
|
Large Language Models (LLMs) offer natural language explanations as an alternative to feature attribution methods for model interpretability. However, despite their plausibility, they may not reflect the model's true reasoning faithfully, which is crucial for understanding the model's true decision-making processes. Although several faithfulness metrics have been proposed, they are often evaluated in isolation, making direct, principled comparisons between them difficult. Here, we present Causal Diagnosticity, a framework that serves as a common testbed to evaluate faithfulness metrics for natural language explanations. Our framework employs the concept of diagnosticity, and uses model-editing methods to generate faithful-unfaithful explanation pairs. Our benchmark includes four tasks: fact-checking, analogy, object counting, and multi-hop reasoning. We evaluate prominent faithfulness metrics, including post-hoc explanation and chain-of-thought-based methods. We find that diagnostic performance varies across tasks and models, with Filler Tokens performing best overall. Additionally, continuous metrics are generally more diagnostic than binary ones but can be sensitive to noise and model choice. Our results highlight the need for more robust faithfulness metrics.
|
[
{
"version": "v1",
"created": "Wed, 26 Feb 2025 05:35:53 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 10:35:38 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Zaman",
"Kerem",
""
],
[
"Srivastava",
"Shashank",
""
]
] |
2502.18950
|
Zhiyun Cheng
|
Zhiyun Cheng
|
Partial-dual genus polynomial of graphs
|
12 pages
|
European Journal of Combinatorics Volume 130, December 2025, 104221
| null | null |
math.CO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Recently, Chmutov introduced the partial duality of ribbon graphs, which can be regarded as a generalization of the classical Euler-Poincar\'e duality. The partial-dual genus polynomial $^\partial\varepsilon_G(z)$ is an enumeration of the partial duals of $G$ by Euler genus. For an intersection graph derived from a given chord diagram, the partial-dual genus polynomial can be defined by considering the ribbon graph associated to the chord diagram. In this paper, we provide a combinatorial approach to the partial-dual genus polynomial in terms of intersection graphs without referring to chord diagrams. After extending the definition of the partial-dual genus polynomial from intersection graphs to all graphs, we prove that it satisfies the four-term relation of graphs. This provides an answer to a problem proposed by Chmutov.
|
[
{
"version": "v1",
"created": "Wed, 26 Feb 2025 08:59:44 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Cheng",
"Zhiyun",
""
]
] |
2502.19137
|
Piotr Sza\'nkowski
|
Piotr Sza\'nkowski
|
A perturbation theory for multi-time correlation functions in open quantum systems
|
Submission to SciPost Physics; updated version: included feedback from reviewers
|
SciPost Phys. 19, 066 (2025)
|
10.21468/SciPostPhys.19.3.066
| null |
quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Dynamical maps are the principal subject of the open system theory. Formally, the dynamical map of a given open quantum system is a density matrix transformation that takes any initial state and sends it to the state at a later time. Physically, it encapsulates the system's evolution due to coupling with its environment. Hence, the theory provides a flexible and accurate framework for computing expectation values of open system observables. However, expectation values -- or more generally, single-time correlation functions -- capture only the simplest aspects of a quantum system's dynamics. A complete characterization of the dynamics requires access to multi-time correlation functions as well: phenomena like detailed balance, linear and non-linear response, non-equilibrium transport in general, or even sequential measurements of system observables are all described in terms of multi-time correlations. For closed systems, such correlations are well-defined, even though knowledge of the system's state alone is insufficient to determine them fully. In contrast, the standard dynamical map formalism for open systems does not account for multi-time correlations, as it is fundamentally limited to describing state evolution. Here, we extend the scope of open quantum system theory by developing a systematic perturbation theory for computing multi-time correlation functions.
|
[
{
"version": "v1",
"created": "Wed, 26 Feb 2025 13:52:13 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Mar 2025 13:44:11 GMT"
},
{
"version": "v3",
"created": "Mon, 04 Aug 2025 12:23:31 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Szańkowski",
"Piotr",
""
]
] |
2502.19155
|
Mohsen Fathi
|
Mohsen Fathi
|
Effect of primary scalar hair on black hole's strong lensing in Beyond Horndeski Gravity
|
23 pages, 14 figures, 3 tables, to appear in Physics of the Dark Universe
| null |
10.1016/j.dark.2025.102069
| null |
gr-qc astro-ph.HE hep-th math-ph math.MP
|
http://creativecommons.org/licenses/by/4.0/
|
We study the gravitational lensing effects of a static, asymptotically flat black hole with primary scalar hair in Beyond Horndeski gravity, focusing on the strong lensing regime. Recently, in Ref. [1], we placed constraints on the scalar hair parameter by analyzing its thermodynamic stability and the black hole shadow. In this work, we further investigate the strong lensing properties of the black hole, which form the basis of shadow formation, and employ observational data from the Event Horizon Telescope to derive more precise constraints on the scalar hair parameter. Additionally, we compute the shape and position of different lensed images of a thin accretion disk and argue that the observed black hole shadow corresponds to the secondary image of the emitting disk. Using this interpretation, we perform a new set of constraints on the scalar hair. Furthermore, we discuss why higher-order images are not suitable for astrophysical constraints, highlighting the limitations posed by their faintness and observational challenges. Finally, we find that models satisfying these constraints exhibit local instabilities.
|
[
{
"version": "v1",
"created": "Wed, 26 Feb 2025 14:13:39 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 14:34:23 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Fathi",
"Mohsen",
""
]
] |
2502.19277
|
Robert N\"urnberg
|
Klaus Deckelnick and Robert N\"urnberg
|
Second order in time finite element schemes for curve shortening flow and curve diffusion
|
23 pages, 2 figures
| null | null | null |
math.NA cs.NA
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
We prove optimal error bounds for a second order in time finite element approximation of curve shortening flow in possibly higher codimension. In addition, we introduce a second order in time method for curve diffusion. Both schemes are based on variational formulations of strictly parabolic systems of partial differential equations that feature a tangential velocity which under discretization is beneficial for the mesh quality. In each time step only two linear systems need to be solved. Numerical experiments demonstrate second order convergence as well as asymptotic equidistribution.
|
[
{
"version": "v1",
"created": "Wed, 26 Feb 2025 16:32:36 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 12:30:59 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Deckelnick",
"Klaus",
""
],
[
"Nürnberg",
"Robert",
""
]
] |
2502.19498
|
H\'ector Sanchis P\'erez
|
H\'ector Sanchis, Gabriela Barenboim, Yuber F. Perez-Gonzalez
|
Neutrino Emission and Plasma Heating from Primordial Black Holes: An Improved Approach to $N_\mathrm{eff}$ Constraints
|
26 pages, 4 figures. Major update. Changes: - Post-emission interactions of emitted unstable particles are now taken into account. - A more recent observational $N_{\mathrm{eff}}$ constraint is used for the bounds. - Various minor changes
|
JCAP 08 (2025) 049
|
10.1088/1475-7516/2025/08/049
| null |
astro-ph.CO hep-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We investigate the impact of neutrino emission via Hawking radiation from primordial black holes (PBHs) on the cosmological effective number of neutrino species, $N_{\mathrm{eff}}$, after neutrino decoupling. By comparing this effect with observational limits, we derive bounds on the abundance of light PBHs. Our analysis incorporates two previously unaccounted-for effects: the emission of secondary neutrinos from unstable particles, which increases $N_{\mathrm{eff}}$, and the modification of the neutrino-photon temperature ratio due to particle emission heating the photon plasma, which lowers $N_{\mathrm{eff}}$. Overall, including these effects allows us to impose constraints on PBHs with initial masses in the range $10^9~{\rm g}\lesssim M_{\rm ini} \lesssim 10^{13}~{\rm g}$. However, our limits remain less stringent than those derived from Big Bang Nucleosynthesis.
|
[
{
"version": "v1",
"created": "Wed, 26 Feb 2025 19:03:23 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 10:29:24 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Sanchis",
"Héctor",
""
],
[
"Barenboim",
"Gabriela",
""
],
[
"Perez-Gonzalez",
"Yuber F.",
""
]
] |
2502.19715
|
Dong Yan
|
Lei Huang, Peng-Fei Wang, Jian-Qi Zhang, Xin Zhou, Shuo Zhang, Han-Xiao Zhang, Hong Yang, Dong Yan
|
Nonreciprocal Entanglement by Dynamically Encircling a Nexus
|
8 pages, 4 figures
| null | null | null |
quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Nonreciprocal entanglement, characterized by inherently robust operation, is a cornerstone for quantum information processing and communications. However, it remains a great challenge to achieve nonreciprocal entanglement characterized by stability and robustness against environmental fluctuations. Here, we propose a universal nonlinear mechanism to engineer magnetic-free nonreciprocity in dissipative optomechanics by utilizing bistability, a phenomenon ubiquitous across nonlinear physical systems. By dynamically encircling the nexus of bistability, a cusp converged by the bistable surfaces, we obtain nonreciprocal displacement and then utilize it to achieve robust nonreciprocal entanglement. Owing to the unique landscape of bistability, our nonreciprocal displacement and entanglements exhibit stability and robustness through closed-loop operations. Our work presents a foundational framework for leveraging nonlinearity to achieve nonreciprocal quantum information processing. It paves new avenues for exploring nonreciprocal quantum information processing and designing backaction-immune quantum metrology with nonlinearity.
|
[
{
"version": "v1",
"created": "Thu, 27 Feb 2025 03:13:26 GMT"
},
{
"version": "v2",
"created": "Tue, 05 Aug 2025 03:02:54 GMT"
},
{
"version": "v3",
"created": "Sat, 30 Aug 2025 15:33:43 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Huang",
"Lei",
""
],
[
"Wang",
"Peng-Fei",
""
],
[
"Zhang",
"Jian-Qi",
""
],
[
"Zhou",
"Xin",
""
],
[
"Zhang",
"Shuo",
""
],
[
"Zhang",
"Han-Xiao",
""
],
[
"Yang",
"Hong",
""
],
[
"Yan",
"Dong",
""
]
] |
2502.20293
|
Zahiriddin Rustamov
|
Zahiriddin Rustamov, Ayham Zaitouny, Nazar Zaki
|
Scalable Graph Attention-based Instance Selection via Mini-Batch Sampling and Hierarchical Hashing
|
Accepted for publication in AI Open
| null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Instance selection (IS) addresses the critical challenge of reducing dataset size while keeping informative characteristics, becoming increasingly important as datasets grow to millions of instances. Current IS methods often struggle with capturing complex relationships in high-dimensional spaces and scale with large datasets. This paper introduces a graph attention-based instance selection (GAIS) method that uses attention mechanisms to identify informative instances through their structural relationships in graph representations. We present two approaches for scalable graph construction: a distance-based mini-batch sampling technique that achieves dataset-size-independent complexity through strategic batch processing, and a hierarchical hashing approach that enables efficient similarity computation through random projections. The mini-batch approach keeps class distributions through stratified sampling, while the hierarchical hashing method captures relationships at multiple granularities through single-level, multi-level, and multi-view variants. Experiments across 39 datasets show that GAIS achieves reduction rates above 96\% while maintaining or improving model performance relative to state-of-the-art IS methods. The findings show that the distance-based mini-batch approach offers an optimal efficiency for large-scale datasets, while multi-view variants excel on complex, high-dimensional data, demonstrating that attention-based importance scoring can effectively identify instances important for maintaining decision boundaries while avoiding computationally prohibitive pairwise comparisons.
|
[
{
"version": "v1",
"created": "Thu, 27 Feb 2025 17:17:53 GMT"
},
{
"version": "v2",
"created": "Tue, 10 Jun 2025 10:12:02 GMT"
},
{
"version": "v3",
"created": "Sun, 31 Aug 2025 19:09:22 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Rustamov",
"Zahiriddin",
""
],
[
"Zaitouny",
"Ayham",
""
],
[
"Zaki",
"Nazar",
""
]
] |
2502.20384
|
Arman Cam
|
\"Ozg\"ur Akarsu, Arman \c{C}am, Evangelos A. Paraskevas, Leandros Perivolaropoulos
|
Linear matter density perturbations in the $\Lambda_{\rm s}$CDM model: Examining growth dynamics and addressing the $S_8$ tension
|
35 pages, 13 figures, 6 tables, matches the version published in JCAP
|
JCAP 08 (2025) 089
|
10.1088/1475-7516/2025/08/089
| null |
astro-ph.CO gr-qc
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We investigate linear matter density perturbations in the $\Lambda_{\rm s}$CDM, in which the $\Lambda$ is replaced by late-time ($z\sim2$) mirror AdS-dS transition, resulting in distinct growth dynamics. We use two complementary approaches: (i) determining the initial density contrast and its evolution rate for a given collapse scale factor, (ii) computing the collapse scale factor for a specified initial density contrast and evolution rate. We derive analytical solutions for the growth rate $f=\Omega_{\rm m}^\gamma$ and growth index $\gamma$ in both models. Prior to the transition, the AdS-like $\Lambda$ reduces cosmic friction, causing linear matter density perturbations to grow more rapidly than in $\Lambda$CDM; this effect is most pronounced just before the transition, with a growth rate around $15\%$ higher than $\Lambda$CDM around $z\sim2$. After the transition, $\Lambda_{\rm s}$CDM behaves similarly to $\Lambda$CDM but features a larger cosmological constant, leading to higher $H(z)$ and greater cosmic friction that more effectively suppresses growth. Before the transition, $\gamma$ remains below both the $\Lambda$CDM and EdS values ($\gamma\approx6/11$); during the transition, it increases rapidly and then grows gradually, paralleling $\Lambda$CDM while remaining slightly higher in the post-transition era-though overall, it stays near $\gamma\sim0.55$. Using the Planck best-fit values, $\Omega_{\rm m0}=0.28$ for $\Lambda_{\mathrm{s}}$CDM and $\Omega_{\rm m0}=0.32$ for $\Lambda$CDM, we find $f=0.49$ and $f=0.53$, respectively. $\Lambda_{\rm s}$CDM predicts a value of $f=0.48$, recently obtained from LSS data when $\gamma$ is treated as a free parameter in $\Lambda$CDM. This suggests that $\Lambda_{\rm s}$CDM may resolve the structure growth anomaly, without deviating from $\gamma \sim 0.55$.
|
[
{
"version": "v1",
"created": "Thu, 27 Feb 2025 18:56:47 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 12:16:33 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Akarsu",
"Özgür",
""
],
[
"Çam",
"Arman",
""
],
[
"Paraskevas",
"Evangelos A.",
""
],
[
"Perivolaropoulos",
"Leandros",
""
]
] |
2502.20396
|
Toru Lin
|
Toru Lin, Kartik Sachdev, Linxi Fan, Jitendra Malik, Yuke Zhu
|
Sim-to-Real Reinforcement Learning for Vision-Based Dexterous Manipulation on Humanoids
|
Published at CoRL 2025. Project page can be found at https://toruowo.github.io/recipe/
| null | null | null |
cs.RO cs.AI cs.CV cs.LG cs.SY eess.SY
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Learning generalizable robot manipulation policies, especially for complex multi-fingered humanoids, remains a significant challenge. Existing approaches primarily rely on extensive data collection and imitation learning, which are expensive, labor-intensive, and difficult to scale. Sim-to-real reinforcement learning (RL) offers a promising alternative, but has mostly succeeded in simpler state-based or single-hand setups. How to effectively extend this to vision-based, contact-rich bimanual manipulation tasks remains an open question. In this paper, we introduce a practical sim-to-real RL recipe that trains a humanoid robot to perform three challenging dexterous manipulation tasks: grasp-and-reach, box lift and bimanual handover. Our method features an automated real-to-sim tuning module, a generalized reward formulation based on contact and object goals, a divide-and-conquer policy distillation framework, and a hybrid object representation strategy with modality-specific augmentation. We demonstrate high success rates on unseen objects and robust, adaptive policy behaviors -- highlighting that vision-based dexterous manipulation via sim-to-real RL is not only viable, but also scalable and broadly applicable to real-world humanoid manipulation tasks.
|
[
{
"version": "v1",
"created": "Thu, 27 Feb 2025 18:59:52 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 23:08:08 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Lin",
"Toru",
""
],
[
"Sachdev",
"Kartik",
""
],
[
"Fan",
"Linxi",
""
],
[
"Malik",
"Jitendra",
""
],
[
"Zhu",
"Yuke",
""
]
] |
2502.21057
|
Taeho Lee
|
Taeho Lee and Donghwan Lee
|
Robust Deterministic Policy Gradient for Disturbance Attenuation and Its Application to Quadrotor Control
|
24 pages
| null | null | null |
cs.RO cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Practical control systems pose significant challenges in identifying optimal control policies due to uncertainties in the system model and external disturbances. While $H_\infty$ control techniques are commonly used to design robust controllers that mitigate the effects of disturbances, these methods often require complex and computationally intensive calculations. To address this issue, this paper proposes a reinforcement learning algorithm called Robust Deterministic Policy Gradient (RDPG), which formulates the $H_\infty$ control problem as a two-player zero-sum dynamic game. In this formulation, one player (the user) aims to minimize the cost, while the other player (the adversary) seeks to maximize it. We then employ deterministic policy gradient (DPG) and its deep reinforcement learning counterpart to train a robust control policy with effective disturbance attenuation. In particular, for practical implementation, we introduce an algorithm called robust deep deterministic policy gradient (RDDPG), which employs a deep neural network architecture and integrates techniques from the twin-delayed deep deterministic policy gradient (TD3) to enhance stability and learning efficiency. To evaluate the proposed algorithm, we implement it on an unmanned aerial vehicle (UAV) tasked with following a predefined path in a disturbance-prone environment. The experimental results demonstrate that the proposed method outperforms other control approaches in terms of robustness against disturbances, enabling precise real-time tracking of moving targets even under severe disturbance conditions.
|
[
{
"version": "v1",
"created": "Fri, 28 Feb 2025 13:58:22 GMT"
},
{
"version": "v2",
"created": "Thu, 06 Mar 2025 11:02:06 GMT"
},
{
"version": "v3",
"created": "Wed, 12 Mar 2025 23:39:47 GMT"
},
{
"version": "v4",
"created": "Tue, 02 Sep 2025 10:25:02 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Lee",
"Taeho",
""
],
[
"Lee",
"Donghwan",
""
]
] |
2503.00059
|
Rui Hu
|
Rui Hu, Delai Qiu, Shuyu Wei, Jiaming Zhang, Yining Wang, Shengping Liu, Jitao Sang
|
Investigating and Enhancing Vision-Audio Capability in Omnimodal Large Language Models
|
Accepted to ACL 2025 Findings
| null | null | null |
cs.CV cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Omnimodal Large Language Models (OLLMs) have shown significant progress in integrating vision and text, but still struggle with integrating vision and audio, often exhibiting suboptimal performance when processing audio queries compared to text queries. This disparity is primarily due to insufficient alignment between vision and audio modalities during training, leading to inadequate attention to visual information when using audio queries. To mitigate this issue, we propose a Self-Knowledge Distillation (Self-KD) training method where the vision-text component of the OLLM serves as the teacher and the vision-audio component as the student. This enables the model to process audio in a manner analogous to its text processing. Our experimental results demonstrate that Self-KD is an effective method for enhancing the vision-audio capabilities of OLLMs by learning from the vision-text components, which subsequently improves the interaction between audio and images and results in improved performance on multimodal tasks.
|
[
{
"version": "v1",
"created": "Thu, 27 Feb 2025 02:19:09 GMT"
},
{
"version": "v2",
"created": "Wed, 21 May 2025 02:59:57 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 09:56:50 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Hu",
"Rui",
""
],
[
"Qiu",
"Delai",
""
],
[
"Wei",
"Shuyu",
""
],
[
"Zhang",
"Jiaming",
""
],
[
"Wang",
"Yining",
""
],
[
"Liu",
"Shengping",
""
],
[
"Sang",
"Jitao",
""
]
] |
2503.00108
|
Subarna Chaki
|
Subarna Chaki (AIfA Bonn & UHH), Andrina Nicola (JBCA & AIfA Bonn), Alessio Spurio Mancini (RHUL), Davide Piras (Geneva) and Robert Reischke (AIfA Bonn)
|
Constraining the primordial power spectrum using a differentiable likelihood
|
31 pages, 15 figures, 2 tables. Version accepted for publication in JCAP
|
JCAP08(2025)068
|
10.1088/1475-7516/2025/08/068
| null |
astro-ph.CO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The simplest inflationary models predict the primordial power spectrum (PPS) of curvature perturbations to be nearly scale-invariant. However, various other models of inflation predict deviations from this behaviour, motivating a data-driven approach to reconstruct the PPS and constrain its shape. In this work, we present a novel method that employs a fully differentiable pipeline to reconstruct the PPS using Gaussian Processes and uses neural network emulators for fast and differentiable theoretical predictions. By leveraging gradient-based sampling techniques, such as Hamiltonian Monte Carlo, our approach efficiently samples the high-dimensional parameter space of cosmological parameters and the free-form PPS, enabling joint constraints on both. Applying this framework to Planck 2018 Cosmic Microwave Background (CMB) temperature anisotropy data we find our reconstructed PPS to be consistent with near scale-invariance on small scales, while exhibiting large uncertainties at large scales, driven mostly by cosmic variance. Our results show an overestimation of the PPS amplitude compared to $\Lambda$CDM predictions from the Planck 2018 analysis, which we attribute to our choice of a wider prior on the optical depth $\tau$ based on Planck 2015 measurements. Adopting a prior consistent with Planck 2018 measurements brings our results into full agreement with previous work. To ensure robustness of our results, we validate our differentiable pipeline against a non-differentiable framework, and also demonstrate that our results are insensitive to the choice of Gaussian process hyperparameters. These promising results and the flexibility of our pipeline make it ideally suited for application to additional data sets such as CMB polarisation as well as Large-Scale Structure probes, thus moving towards multi-probe primordial power spectrum reconstruction.
|
[
{
"version": "v1",
"created": "Fri, 28 Feb 2025 19:00:01 GMT"
},
{
"version": "v2",
"created": "Sun, 27 Jul 2025 11:03:58 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Chaki",
"Subarna",
"",
"AIfA Bonn & UHH"
],
[
"Nicola",
"Andrina",
"",
"JBCA & AIfA Bonn"
],
[
"Mancini",
"Alessio Spurio",
"",
"RHUL"
],
[
"Piras",
"Davide",
"",
"Geneva"
],
[
"Reischke",
"Robert",
"",
"AIfA Bonn"
]
] |
2503.00134
|
Zhongqi Yang
|
Zhongqi Yang and Amir Rahmani
|
Personalized Causal Graph Reasoning for LLMs: An Implementation for Dietary Recommendations
| null | null | null | null |
cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Large Language Models (LLMs) excel at general-purpose reasoning by leveraging broad commonsense knowledge, but they remain limited in tasks requiring personalized reasoning over multifactorial personal data. This limitation constrains their applicability in domains such as healthcare, where decisions must adapt to individual contexts. We introduce Personalized Causal Graph Reasoning, a framework that enables LLMs to reason over individual-specific causal graphs constructed from longitudinal data. Each graph encodes how user-specific factors influence targeted outcomes. In response to a query, the LLM traverses the graph to identify relevant causal pathways, rank them by estimated impact, simulate potential outcomes, and generate tailored responses. We implement this framework in the context of nutrient-oriented dietary recommendations, where variability in metabolic responses demands personalized reasoning. Using counterfactual evaluation, we assess the effectiveness of LLM-generated food suggestions for glucose control. Our method reduces postprandial glucose iAUC across three time windows compared to prior approaches. Additional LLM-as-a-judge evaluations further confirm improvements in personalization quality.
|
[
{
"version": "v1",
"created": "Fri, 28 Feb 2025 19:25:04 GMT"
},
{
"version": "v2",
"created": "Wed, 28 May 2025 17:18:49 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 04:16:57 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Yang",
"Zhongqi",
""
],
[
"Rahmani",
"Amir",
""
]
] |
2503.00169
|
Dip Joti Paul
|
Dip Joti Paul, Tony X. Zhou, Karl K. Berggren
|
Determination of Mid-Infrared Refractive Indices of Superconducting Thin Films Using Fourier Transform Infrared Spectroscopy
|
9 figures
|
Appl. Phys. Lett. 126, 252601 (2025)
|
10.1063/5.0268308
| null |
physics.optics cond-mat.supr-con
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
In this work, we present a technique to determine the mid-infrared refractive indices of thin superconducting films using Fourier transform infrared spectroscopy (FTIR). In particular, we performed FTIR transmission and reflection measurements on 10-nm-thick NbN and 15-nm-thick MoSi films in the wavelength range of 2.5 to 25 $\mu$m, corresponding to frequencies of 12-120 THz or photon energies of 50-500 meV. To extract the mid-infrared refractive indices of these thin films, we used the Drude-Lorentz oscillator model to represent their dielectric functions and implemented an optimization algorithm to fit the oscillator parameters by minimizing the error between the measured and simulated FTIR spectra. We performed Monte Carlo simulations in the optimization routine to estimate error ranges in the extracted refractive indices resulting from multiple sources of measurement uncertainty. To evaluate the consistency of the extracted dielectric functions, we compared the refractive indices extrapolated from these dielectric functions in the UV to near-infrared wavelengths with the values separately measured using spectroscopic ellipsometry. We validated the applicability of the extracted mid-infrared refractive indices of NbN and MoSi at temperatures below their critical temperatures by comparing them with the Mattis-Bardeen model. This FTIR-based refractive index measurement approach can be extended to measure the refractive indices of thin films at wavelengths beyond 25 $\mu$m, which will be useful for designing highly efficient photon detectors and photonic devices with enhanced optical absorption in the mid- and far-infrared wavelengths.
|
[
{
"version": "v1",
"created": "Fri, 28 Feb 2025 20:31:21 GMT"
},
{
"version": "v2",
"created": "Fri, 16 May 2025 18:33:02 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Paul",
"Dip Joti",
""
],
[
"Zhou",
"Tony X.",
""
],
[
"Berggren",
"Karl K.",
""
]
] |
2503.00229
|
Sharan Vaswani
|
Sharan Vaswani, Reza Babanezhad
|
Armijo Line-search Can Make (Stochastic) Gradient Descent Provably Faster
|
ICML 2025. 37 pages
| null | null | null |
cs.LG math.OC stat.ML
|
http://creativecommons.org/licenses/by/4.0/
|
Armijo line-search (Armijo-LS) is a standard method to set the step-size for gradient descent (GD). For smooth functions, Armijo-LS alleviates the need to know the global smoothness constant L and adapts to the ``local'' smoothness, enabling GD to converge faster. Existing theoretical analyses show that GD with Armijo-LS (GD-LS) can result in constant factor improvements over GD with a 1/L step-size (denoted as GD(1/L)). We strengthen these results and show that if the objective function satisfies a certain non-uniform smoothness condition, GD-LS can result in a faster convergence rate than GD(1/L). In particular, we prove that for convex objectives corresponding to logistic regression and multi-class classification, GD-LS can converge to the optimum at a linear rate, and hence improves over the sublinear convergence of GD(1/L). Furthermore, for non-convex objectives satisfying gradient domination (e.g., those corresponding to the softmax policy gradient in RL or generalized linear models with a logistic link function), GD-LS can match the fast convergence of algorithms tailored for these specific settings. Finally, we analyze the convergence of stochastic GD with a stochastic line-search on convex losses under the interpolation assumption.
|
[
{
"version": "v1",
"created": "Fri, 28 Feb 2025 22:26:33 GMT"
},
{
"version": "v2",
"created": "Tue, 03 Jun 2025 14:29:26 GMT"
},
{
"version": "v3",
"created": "Tue, 02 Sep 2025 02:33:29 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Vaswani",
"Sharan",
""
],
[
"Babanezhad",
"Reza",
""
]
] |
2503.00271
|
Kelechi Gabriel Kalu
|
Kelechi G. Kalu, Sofia Okorafor, Tanmay Singla, Sophie Chen, Santiago Torres-Arias, James C. Davis
|
Why Johnny Signs with Next-Generation Tools: A Usability Case Study of Sigstore
|
25 Pages
| null | null | null |
cs.SE cs.CR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Software signing is the most robust method for ensuring the integrity and authenticity of components in a software supply chain. Traditional signing tools burdened practitioners with key management and signer identification, creating both usability challenges and security risks. A new class of next-generation signing tools has automated many of these concerns, but little is known about their usability and its effect on adoption and effectiveness in practice. A usability evaluation can clarify the extent to which next-generation designs succeed and highlight priorities for improvement.
To fill this gap, we conducted a usability study of Sigstore, a pioneering and widely adopted exemplar of next-generation signing. Through interviews with 17 industry experts, we examined (1) the problems and advantages associated with practitioners' tooling choices, (2) how and why their signing-tool usage has evolved over time, and (3) the contexts that cause usability concerns. Our findings illuminate the usability factors of next-generation signing tools and yield recommendations for toolmakers, adopting organizations, and the research community. Notably, components of next-generation tooling exhibit different levels of maturity and readiness for adoption, and integration flexibility is a common pain point, but potentially mitigable through plugins and APIs. Our results will help next-generation signing toolmakers further strengthen software supply chain security.
|
[
{
"version": "v1",
"created": "Sat, 01 Mar 2025 00:59:18 GMT"
},
{
"version": "v2",
"created": "Tue, 04 Mar 2025 19:55:28 GMT"
},
{
"version": "v3",
"created": "Fri, 14 Mar 2025 00:30:15 GMT"
},
{
"version": "v4",
"created": "Wed, 04 Jun 2025 22:10:22 GMT"
},
{
"version": "v5",
"created": "Fri, 29 Aug 2025 20:16:05 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Kalu",
"Kelechi G.",
""
],
[
"Okorafor",
"Sofia",
""
],
[
"Singla",
"Tanmay",
""
],
[
"Chen",
"Sophie",
""
],
[
"Torres-Arias",
"Santiago",
""
],
[
"Davis",
"James C.",
""
]
] |
2503.00438
|
Suhyoung Choi
|
Suhyoung Choi
|
Real projective orbifolds with ends and their deformation theory: The deformation theory for nicest ones
|
xiii + 429 pages, 7 figures
| null | null | null |
math.GT math.DG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This monograph is on convex real projective structures on strongly tame n-orbifolds with some appropriate conditions on ends.
|
[
{
"version": "v1",
"created": "Sat, 01 Mar 2025 10:32:03 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 12:30:45 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Choi",
"Suhyoung",
""
]
] |
2503.00600
|
Alexander Lee
|
Alexander W. Lee, Justin Chan, Michael Fu, Nicolas Kim, Akshay Mehta, Deepti Raghavan, Ugur Cetintemel
|
Semantic Integrity Constraints: Declarative Guardrails for AI-Augmented Data Processing Systems
| null |
PVLDB, 18(11): 4073 - 4080, 2025
|
10.14778/3749646.3749677
| null |
cs.DB cs.AI cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
AI-augmented data processing systems (DPSs) integrate large language models (LLMs) into query pipelines, allowing powerful semantic operations on structured and unstructured data. However, the reliability (a.k.a. trust) of these systems is fundamentally challenged by the potential for LLMs to produce errors, limiting their adoption in critical domains. To help address this reliability bottleneck, we introduce semantic integrity constraints (SICs) -- a declarative abstraction for specifying and enforcing correctness conditions over LLM outputs in semantic queries. SICs generalize traditional database integrity constraints to semantic settings, supporting common types of constraints, such as grounding, soundness, and exclusion, with both reactive and proactive enforcement strategies.
We argue that SICs provide a foundation for building reliable and auditable AI-augmented data systems. Specifically, we present a system design for integrating SICs into query planning and runtime execution and discuss its realization in AI-augmented DPSs. To guide and evaluate our vision, we outline several design goals -- covering criteria around expressiveness, runtime semantics, integration, performance, and enterprise-scale applicability -- and discuss how our framework addresses each, along with open research challenges.
|
[
{
"version": "v1",
"created": "Sat, 01 Mar 2025 19:59:25 GMT"
},
{
"version": "v2",
"created": "Mon, 02 Jun 2025 02:31:38 GMT"
},
{
"version": "v3",
"created": "Thu, 07 Aug 2025 03:48:22 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Lee",
"Alexander W.",
""
],
[
"Chan",
"Justin",
""
],
[
"Fu",
"Michael",
""
],
[
"Kim",
"Nicolas",
""
],
[
"Mehta",
"Akshay",
""
],
[
"Raghavan",
"Deepti",
""
],
[
"Cetintemel",
"Ugur",
""
]
] |
2503.01000
|
Karlo Lukic
|
Karlo Lukic and Lazaros Papadopoulos
|
Privacy vs. Profit: The Impact of Google's Manifest Version 3 (MV3) Update on Ad Blocker Effectiveness
|
24 pages, 11 figures. Accepted to Proceedings on Privacy Enhancing Technologies (PoPETs), Volume 2026, Issue 1
| null | null | null |
cs.CY econ.GN q-fin.EC
|
http://creativecommons.org/licenses/by/4.0/
|
Google's recent update to the manifest file for Chrome browser extensions, transitioning from manifest version 2 (MV2) to manifest version 3 (MV3), has raised concerns among users and ad blocker providers, who worry that the new restrictions, notably the shift from the powerful WebRequest API to the more restrictive DeclarativeNetRequest API, might reduce ad blocker effectiveness. Because ad blockers play a vital role for millions of users seeking a more private and ad-free browsing experience, this study empirically investigates how the MV3 update affects their ability to block ads and trackers. Through a browser-based experiment conducted across multiple samples of ad-supported websites, we compare the MV3 to MV2 instances of four widely used ad blockers. Our results reveal no statistically significant reduction in ad-blocking or anti-tracking effectiveness for MV3 ad blockers compared to their MV2 counterparts, and in some cases, MV3 instances even exhibit slight improvements in blocking trackers. These findings are reassuring for users, indicating that the MV3 instances of popular ad blockers continue to provide effective protection against intrusive ads and privacy-infringing trackers. While some uncertainties remain, ad blocker providers appear to have successfully navigated the MV3 update, finding solutions that maintain the core functionality of their extensions.
|
[
{
"version": "v1",
"created": "Sun, 02 Mar 2025 19:41:34 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 04:41:20 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Lukic",
"Karlo",
""
],
[
"Papadopoulos",
"Lazaros",
""
]
] |
2503.01312
|
Carlos Vivo-Vilches
|
Carlos Vivo-Vilches, Esad Hrnjic, Martin Martschini, Kyra Altindag, Lee William Packer, Robin Golser, Karin Hain
|
Towards AMS measurements of $^{91}$Nb, $^{94}$Nb and $^{93}$Mo produced in fusion environment
|
14 pages, 6 figures, 1 table
|
Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms, Volume 568, November 2025, 165847
|
10.1016/j.nimb.2025.165847
| null |
physics.ins-det nucl-ex
|
http://creativecommons.org/licenses/by/4.0/
|
Long-lived radionuclides, such as $^{91}$Nb, $^{94}$Nb and $^{93}$Mo, are expected to be produced in nuclear fusion reactors by reactions of high-energy neutrons with the structural material. Accurate predictions of waste categorization require experimental validation of simulation codes like FISPACT-II. This work explores the use of Ion-Laser InterAction Mass Spectrometry (ILIAMS) at the Vienna Environmental Research Accelerator (VERA) to measure these three radionuclides by accelerator mass spectrometry (AMS). The ILIAMS setup employs laser photodetachment to suppress their respective stable isobars: $^{91}$Zr, $^{94}$Zr and $^{94}$Mo, and $^{93}$Nb. For $^{91,94}$Nb measurements, NbO$_3^-$ is selected, with interferences from ZrO$_3^-$ suppressed by collisions with the He buffer gas in the ion cooler. The suppression can be enhanced by overlapping a 355\,nm laser with the ion beam. The lower limit for the suppression factor is 37000. In that way, we reach $^{91}$Zr/$^{93}$Nb and $^{94}$Zr/$^{93}$Nb levels of $1.2\,\times\,10^{-14}$ and $1.8\,\times\,10^{-14}$, respectively, in targets prepared from commercial Nb$_2$O$_5$. MoO$_3$ is suppressed by a factor of 4360, leading to a $^{94}$Mo/$^{93}$Nb interference of $1.28\,\times\,10^{-10}$ in the same targets. For $^{93}$Mo measurements, MoO$_2^-$ is selected, with interference from $^{93}$NbO$_2^-$ suppressed by 637\,nm photons by a factor of $5.5\,\times\,10^6$. This results in a $^{93}$Nb/$^\text{nat}$Mo level of $1.3\,\times\,10^{-13}$ in targets prepared from commercial MoO$_3$. Suppression factors as high as this are not achieved by isobar suppression techniques based on differences in energy loss, not even by AMS facilities with terminal voltages above 8.5\,MV.
|
[
{
"version": "v1",
"created": "Mon, 03 Mar 2025 08:56:17 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 12:37:01 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Vivo-Vilches",
"Carlos",
""
],
[
"Hrnjic",
"Esad",
""
],
[
"Martschini",
"Martin",
""
],
[
"Altindag",
"Kyra",
""
],
[
"Packer",
"Lee William",
""
],
[
"Golser",
"Robin",
""
],
[
"Hain",
"Karin",
""
]
] |
2503.01392
|
Thomas Walpuski
|
Gorapada Bera and Thomas Walpuski
|
Dirac operators twisted by ramified Euclidean line bundles
| null | null | null | null |
math.DG
|
http://creativecommons.org/licenses/by/4.0/
|
This article is concerned with the analysis of Dirac operators $D$ twisted by ramified Euclidean line bundles $(Z,\mathfrak{l})$-motivated by their relation with harmonic $\mathbf{Z}/2\mathbf{Z}$ spinors, which have appeared in various context in gauge theory and calibrated geometry. The closed extensions of $D$ are described in terms of the Gelfand-Robbin quotient $\check{\mathbf{H}}$. Assuming that the branching locus $Z$ is a closed cooriented codimension two submanifold, a geometric realisation of $\check{\mathbf{H}}$ is constructed. This, in turn, leads to an $L^2$ regularity theory.
|
[
{
"version": "v1",
"created": "Mon, 03 Mar 2025 10:46:08 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 08:43:15 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Bera",
"Gorapada",
""
],
[
"Walpuski",
"Thomas",
""
]
] |
2503.01400
|
Tomasz Rybotycki
|
Dawid Mazur, Tomasz Rybotycki and Piotr Gawron
|
Hyperspectral image segmentation with a machine learning model trained using quantum annealer
|
16 pages, 6 figures, 3 tables
|
Lecture Notes in Computer Science, vol 15911, 2025
|
10.1007/978-3-031-97570-7_16
| null |
quant-ph cs.CV cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Training of machine learning models consumes large amounts of energy. Since the energy consumption becomes a major problem in the development and implementation of artificial intelligence systems there exists a need to investigate the ways to reduce use of the resources by these systems. In this work we study how application of quantum annealers could lead to reduction of energy cost in training models aiming at pixel-level segmentation of hyperspectral images. Following the results of QBM4EO team, we propose a classical machine learning model, partially trained using quantum annealer, for hyperspectral image segmentation. We show that the model trained using quantum annealer is better or at least comparable with models trained using alternative algorithms, according to the preselected, common metrics. While direct energy use comparison does not make sense at the current stage of quantum computing technology development, we believe that our work proves that quantum annealing should be considered as a tool for training at least some machine learning models.
|
[
{
"version": "v1",
"created": "Mon, 03 Mar 2025 10:57:06 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Mazur",
"Dawid",
""
],
[
"Rybotycki",
"Tomasz",
""
],
[
"Gawron",
"Piotr",
""
]
] |
2503.01833
|
Roberto Rosati
|
Roberto Rosati, Sai Shradha, Julian Picker, Andrey Turchanin, Bernhard Urbaszek, Ermin Malic
|
Impact of charge transfer excitons on unidirectional exciton transport in lateral TMD heterostructures
|
7 pages, 4 figures; Supplementary Information
|
Nano Lett. 2025, 25, 29, 11319
|
10.1021/acs.nanolett.5c02447
| null |
cond-mat.mes-hall
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Lateral heterostructures built of monolayers of transition metal dichalcogenides (TMDs) are characterized by a thin 1D interface exhibiting a large energy offset. Recently, the formation of spatially separated charge-transfer (CT) excitons at the interface has been demonstrated. The technologically important exciton propagation across the interface and the impact of CT excitons has remained in the dark so far. In this work, we microscopically investigate the spatiotemporal exciton dynamics in the exemplary hBN-encapsulated WSe$_2$-MoSe$_2$ lateral heterostructure and reveal a highly interesting interplay of energy offset-driven unidirectional exciton drift across the interface and efficient capture into energetically lower CT excitons at the interface. This interplay triggers a counterintuitive thermal control of exciton transport with a less efficient propagation at lower temperatures - opposite to the behavior in conventional semiconductors. We predict clear signatures of this intriguing exciton propagation both in far- and near-field photoluminescence experiments. Our results present an important step toward a microscopic understanding of the technologically relevant unidirectional exciton transport in lateral heterostructures.
|
[
{
"version": "v1",
"created": "Mon, 03 Mar 2025 18:55:14 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Rosati",
"Roberto",
""
],
[
"Shradha",
"Sai",
""
],
[
"Picker",
"Julian",
""
],
[
"Turchanin",
"Andrey",
""
],
[
"Urbaszek",
"Bernhard",
""
],
[
"Malic",
"Ermin",
""
]
] |
2503.02131
|
Arya Akhavan
|
Arya Akhavan, Alexandre B. Tsybakov
|
Gradient-free stochastic optimization for additive models
| null | null | null | null |
stat.ML cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
We address the problem of zero-order optimization from noisy observations for an objective function satisfying the Polyak-{\L}ojasiewicz or the strong convexity condition. Additionally, we assume that the objective function has an additive structure and satisfies a higher-order smoothness property, characterized by the H\"older family of functions. The additive model for H\"older classes of functions is well-studied in the literature on nonparametric function estimation, where it is shown that such a model benefits from a substantial improvement of the estimation accuracy compared to the H\"older model without additive structure. We study this established framework in the context of gradient-free optimization. We propose a randomized gradient estimator that, when plugged into a gradient descent algorithm, allows one to achieve minimax optimal optimization error of the order $dT^{-(\beta-1)/\beta}$, where $d$ is the dimension of the problem, $T$ is the number of queries and $\beta\ge 2$ is the H\"older degree of smoothness. We conclude that, in contrast to nonparametric estimation problems, no substantial gain of accuracy can be achieved when using additive models in gradient-free optimization.
|
[
{
"version": "v1",
"created": "Mon, 03 Mar 2025 23:39:08 GMT"
},
{
"version": "v2",
"created": "Sat, 05 Apr 2025 13:02:50 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 10:15:29 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Akhavan",
"Arya",
""
],
[
"Tsybakov",
"Alexandre B.",
""
]
] |
2503.02482
|
Yufei Zhang
|
Chenyu Wu, Shaoguang Zhang, Changxin Guo, Yufei Zhang
|
Data-driven Turbulence Modeling for Separated Flows Considering Non-Local Effect
|
42 pages, 35 figures
|
AIAA Journal 2025
|
10.2514/1.J065567
| null |
physics.flu-dyn
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This study aims to enhance the generalizability of Reynolds-averaged Navier-Stokes (RANS) turbulence models, which are crucial for engineering applications. Classic RANS turbulence models often struggle to predict separated flows accurately. Recently, Data-driven machine learning approaches for turbulence modeling have been explored to address this issue. However, these models are often criticized for their limited generalizability. In this study, we address this issue by incorporating non-local effects into data-driven turbulence modeling. Specifically, we introduce a transport equation for the correction term \b{eta} of the shear stress transport (SST) model to encode non-local information along the mean streamline. The coefficients of the equation are calibrated using high-fidelity data from the NASA hump and periodic hills. The resulting model, termed the \b{eta}-Transport model, demonstrates high accuracy across various separated flows outside the training set, including periodic hills with different geometries and Reynolds numbers, the curved backward-facing step, a two-dimensional bump, and a three-dimensional simplified car body. In all tested cases, the \b{eta}-Transport model yields smaller or equal prediction errors compared to those from classic local data-driven turbulence models using algebraic relations for \b{eta}. These results indicate improved generalizability for separated flows. Furthermore, the \b{eta}-Transport model shows similar accuracy to the baseline model in basic flows including the mixing layer and the channel flow. Therefore, the non-local modeling approach presented here offers a promising pathway for developing more generalizable data-driven turbulence models.
|
[
{
"version": "v1",
"created": "Tue, 04 Mar 2025 10:47:26 GMT"
},
{
"version": "v2",
"created": "Wed, 05 Mar 2025 01:53:53 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Wu",
"Chenyu",
""
],
[
"Zhang",
"Shaoguang",
""
],
[
"Guo",
"Changxin",
""
],
[
"Zhang",
"Yufei",
""
]
] |
2503.02818
|
Michael Howes
|
Persi Diaconis, Michael Howes
|
Random sampling of contingency tables and partitions: Two practical examples of the Burnside process
|
Replaced with version published in Statistics and Computing. 25 pages, 7 figures,
|
Stat Comput 35, 181 (2025)
|
10.1007/s11222-025-10708-5
| null |
stat.CO math.CO math.PR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper gives new, efficient algorithms for approximate uniform sampling of contingency tables and integer partitions. The algorithms use the Burnside process, a general algorithm for sampling a uniform orbit of a finite group acting on a finite set. We show that a technique called `lumping' can be used to derive efficient implementations of the Burnside process. For both contingency tables and partitions, the lumped processes have far lower per step complexity than the original Markov chains. We also define a second Markov chain for partitions called the reflected Burnside process. The reflected Burnside process maintains the computational advantages of the lumped process but empirically converges to the uniform distribution much more rapidly. By using the reflected Burnside process we can easily sample uniform partitions of size $10^{10}$.
|
[
{
"version": "v1",
"created": "Tue, 04 Mar 2025 17:46:07 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 18:22:23 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Diaconis",
"Persi",
""
],
[
"Howes",
"Michael",
""
]
] |
2503.02925
|
Sahand Seifnashri
|
Sahand Seifnashri, Shu-Heng Shao, Xinping Yang
|
Gauging non-invertible symmetries on the lattice
|
66 pages, 1 figure, 1 table; v2: minor changes
|
SciPost Phys. 19, 063 (2025)
|
10.21468/SciPostPhys.19.2.063
|
MIT-CTP/5842, YITP-SB-2025-03
|
cond-mat.str-el hep-th quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We provide a general prescription for gauging finite non-invertible symmetries in 1+1d lattice Hamiltonian systems. Our primary example is the Rep(D$_8$) fusion category generated by the Kennedy-Tasaki transformation, which is the simplest anomaly-free non-invertible symmetry on a spin chain of qubits. We explicitly compute its lattice F-symbols and illustrate our prescription for a particular (non-maximal) gauging of this symmetry. In our gauging procedure, we introduce two qubits around each link, playing the role of "gauge fields" for the non-invertible symmetry, and impose novel Gauss's laws. Similar to the Kramers-Wannier transformation for gauging an ordinary $\mathbb{Z}_2$, our gauging can be summarized by a gauging map, which is part of a larger, continuous non-invertible cosine symmetry.
|
[
{
"version": "v1",
"created": "Tue, 04 Mar 2025 19:00:00 GMT"
},
{
"version": "v2",
"created": "Sat, 19 Jul 2025 19:50:25 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Seifnashri",
"Sahand",
""
],
[
"Shao",
"Shu-Heng",
""
],
[
"Yang",
"Xinping",
""
]
] |
2503.03198
|
Ziqian Yao
|
Ziqian Yao, Clarissa Daniel, Eric Stolt, Vakhtang Chulukhadze, Juan Rivas Davila, Ruochen Lu
|
Lithium Tantalate Bulk Acoustic Resonator For Piezoelectric Power Conversion
|
Submitted to The 23rd International Conference on Solid-State Sensors, Actuators and Microsystems (Transducers 2025)
|
10.1109/Transducers61432.2025.11110984
|
10.1109/Transducers61432.2025.11110984
| null |
physics.app-ph
|
http://creativecommons.org/licenses/by/4.0/
|
We present the first lithium tantalate (LT) thickness-extensional (TE) mode bulk acoustic resonators designed for piezoelectric power conversion, showcasing a low temperature coefficient of frequency (TCF) of -13.56 ppm/K. These resonators also exhibit high quality factors (Q) of 1698, and electromechanical coupling coefficients ($k^2$) of 8.8%, making them suitable for efficient power conversion applications. A grounded ring structure is leveraged for spurious mode-free response and figure of merit (FoM=$k^2 \times Q$) enhancement near the series resonance. The temperature dependency of Q and $k^2$ is experimentally tested over a wide temperature range, from $25^{\circ}\text{C}$ to $130^{\circ}\text{C}$, demonstrating the resonators thermal stability and consistent performance under varying conditions. This work highlights the potential of LT resonators in the future development of thermally stable power electronic systems, enabling more reliable and efficient piezoelectric power converters.
|
[
{
"version": "v1",
"created": "Wed, 05 Mar 2025 05:30:58 GMT"
},
{
"version": "v2",
"created": "Tue, 29 Apr 2025 20:16:16 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Yao",
"Ziqian",
""
],
[
"Daniel",
"Clarissa",
""
],
[
"Stolt",
"Eric",
""
],
[
"Chulukhadze",
"Vakhtang",
""
],
[
"Davila",
"Juan Rivas",
""
],
[
"Lu",
"Ruochen",
""
]
] |
2503.03475
|
Linyu Fan
|
Linyu Fan, Che Wang, Ming Ye, Qizhi Yang, Zejun Wu, Xinghao Ding, Yue Huang, Jianfeng Bao, Shuhui Cai, Congbo Cai
|
Bridging Synthetic-to-Real Gaps: Frequency-Aware Perturbation and Selection for Single-shot Multi-Parametric Mapping Reconstruction
|
This paper is in a very niche field. We need to improve it to make it more applicable to all quantitative MRIs
| null | null | null |
eess.IV cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Data-centric artificial intelligence (AI) has remarkably advanced medical imaging, with emerging methods using synthetic data to address data scarcity while introducing synthetic-to-real gaps. Unsupervised domain adaptation (UDA) shows promise in ground truth-scarce tasks, but its application in reconstruction remains underexplored. Although multiple overlapping-echo detachment (MOLED) achieves ultra-fast multi-parametric reconstruction, extending its application to various clinical scenarios, the quality suffers from deficiency in mitigating the domain gap, difficulty in maintaining structural integrity, and inadequacy in ensuring mapping accuracy. To resolve these issues, we proposed frequency-aware perturbation and selection (FPS), comprising Wasserstein distance-modulated frequency-aware perturbation (WDFP) and hierarchical frequency-aware selection network (HFSNet), which integrates frequency-aware adaptive selection (FAS), compact FAS (cFAS) and feature-aware architecture integration (FAI). Specifically, perturbation activates domain-invariant feature learning within uncertainty, while selection refines optimal solutions within perturbation, establishing a robust and closed-loop learning pathway. Extensive experiments on synthetic data, along with diverse real clinical cases from 5 healthy volunteers, 94 ischemic stroke patients, and 46 meningioma patients, demonstrate the superiority and clinical applicability of FPS. Furthermore, FPS is applied to diffusion tensor imaging (DTI), underscoring its versatility and potential for broader medical applications. The code is available at https://github.com/flyannie/FPS.
|
[
{
"version": "v1",
"created": "Wed, 05 Mar 2025 13:10:11 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 03:11:40 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Fan",
"Linyu",
""
],
[
"Wang",
"Che",
""
],
[
"Ye",
"Ming",
""
],
[
"Yang",
"Qizhi",
""
],
[
"Wu",
"Zejun",
""
],
[
"Ding",
"Xinghao",
""
],
[
"Huang",
"Yue",
""
],
[
"Bao",
"Jianfeng",
""
],
[
"Cai",
"Shuhui",
""
],
[
"Cai",
"Congbo",
""
]
] |
2503.03674
|
Xinze Yang Mr
|
Xinze Yang, Alexander F. Kemper, Adrian Gozar, Eduardo H. da Silva Neto
|
Intermediate band analysis in Green's functions calculations of quasiparticle interference
| null | null |
10.1103/wxj7-3tyj
| null |
cond-mat.str-el
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The measurement of quasiparticle scattering patterns on material surfaces using scanning tunneling microscopy (STM) is now an established technique for accessing the momentum-resolved electronic band structure of solids. However, since these quasiparticle interference (QPI) patterns reflect spatial variations related to differences in the band momenta rather than the momenta themselves, their interpretation often relies on comparisons with simple geometrical models such as the joint density of states (JDOS) or with the convolution of Green's functions. In this paper, we highlight non-intuitive differences between Green's function and JDOS results. To understand the origin of these discrepancies, we analyze the convolution of Green's functions using the Feynman parametrization technique and introduce a framework that we call the intermediate band analysis. This approach allows us to derive simple selection rules for interband QPI, based on electron group velocities. Connecting the intermediate band analysis with the experiment, we consider experimental Bogoliubov QPI patterns measured for FeSe1-xSx, which were recently used to demonstrate a highly anisotropic superconducting gap, indicating superconductivity mediated by nematic fluctuations [1]. The calculated Green's functions convolutions reproduce the particle-hole asymmetry in the intensity of QPI patterns across the Fermi level observed in experiments. Finally, we demonstrate the utility of intermediate band analysis in tracing the origin of this asymmetry to a coherence factor effect of the superconducting state.
|
[
{
"version": "v1",
"created": "Wed, 05 Mar 2025 17:10:22 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Yang",
"Xinze",
""
],
[
"Kemper",
"Alexander F.",
""
],
[
"Gozar",
"Adrian",
""
],
[
"Neto",
"Eduardo H. da Silva",
""
]
] |
2503.03700
|
Niklas Witt
|
Niklas Witt, Siheon Ryee, Lennart Klebl, Jennifer Cano, Giorgio Sangiovanni, Tim O. Wehling
|
Quantum geometry and local moment swapover in correlated graphene heterostructures
|
10 pages, 4 figures + supplement with 11 pages, 9 figures
| null | null | null |
cond-mat.str-el cond-mat.mes-hall
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Graphene-based multilayer systems serve as versatile platforms for exploring the interplay between electron correlation and topology, thanks to distinctive low-energy bands marked by significant quantum metric and Berry curvature from graphene's Dirac bands. Here, we investigate Mott physics and local spin moments in Dirac bands hybridized with a flat band of localized orbitals in functionalized graphene. Via hybridization control, a topological transition is realized between two symmetry-distinct site-selective Mott states featuring local moments in different Wyckoff positions, with a geometrically enforced metallic state emerging in between. We find that this geometrically controlled local moment ``swapover'' and associated metal-insulator physics may be realized through proximity coupling of epitaxial graphene on SiC(0001) with group IV intercalants, where the Mott state faces geometrical obstruction in the large-hybridization limit. Our work shows that chemically functionalized graphene provides a correlated electron platform, very similar to the topological heavy fermions in graphene moir\'e systems but at significantly enhanced characteristic energy scales.
|
[
{
"version": "v1",
"created": "Wed, 05 Mar 2025 17:52:20 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 13:37:46 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Witt",
"Niklas",
""
],
[
"Ryee",
"Siheon",
""
],
[
"Klebl",
"Lennart",
""
],
[
"Cano",
"Jennifer",
""
],
[
"Sangiovanni",
"Giorgio",
""
],
[
"Wehling",
"Tim O.",
""
]
] |
2503.04172
|
Izidor Benedi\v{c}i\v{c}
|
Izidor Benedi\v{c}i\v{c}, Yuri Tanuma, \v{Z}iga Gosar, Bastien An\'ezo, Mariusz Mr\'ozek, Adam Wojciechowski, Denis Ar\v{c}on
|
Spin-lattice relaxation of NV centers in nanodiamonds adsorbed on conducting and non-conducting surfaces
|
13 pages, 5 figures
|
Phys. Rev. B 111, 235421 (2025)
|
10.1103/1s53-5wls
| null |
cond-mat.mes-hall
|
http://creativecommons.org/licenses/by/4.0/
|
The nitrogen-vacancy (NV) centers in nanodiamonds can be utilized as low-cost, highly versatile quantum sensors for studying surface properties in condensed matter physics through the application of relaxometry protocols. For such applications, a detailed knowledge of the intrinsic relaxation processes of NV centers in nanodiamonds is necessary. Here, we study the spin-lattice relaxation rates of NV ensembles in nanodiamonds with average diameters of 40 nm and 3 $\mu$m between room temperature and $\sim$ 6 K. The NV relaxation curves fit to a stretched-exponential form with a stretching exponent $\alpha \approx 0.7$, implying the large distribution of relaxation times of individual centers within nanodiamonds. We determine the Orbach-like scattering on phonons as the leading relaxation mechanism. Finally, we discuss the viability of nanodiamonds as surface sensors when deposited on a metallic substrate and emphasize the need for well-controlled surface preparation techniques.
|
[
{
"version": "v1",
"created": "Thu, 06 Mar 2025 07:37:04 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 13:06:37 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Benedičič",
"Izidor",
""
],
[
"Tanuma",
"Yuri",
""
],
[
"Gosar",
"Žiga",
""
],
[
"Anézo",
"Bastien",
""
],
[
"Mrózek",
"Mariusz",
""
],
[
"Wojciechowski",
"Adam",
""
],
[
"Arčon",
"Denis",
""
]
] |
2503.05011
|
Renjie Li
|
Renjie Li, Weiguang Cui, Ang Liu, Huiyuan Wang, Atulit Srivastava, Romeel Dave, Frazer R. Pearce
|
The Three Hundred: Gas Properties Outside of Galaxy Cluster with the WHIM Contribution and Detection
|
9 pages, 8 figures
|
A&A 701, A37 (2025)
|
10.1051/0004-6361/202554390
| null |
astro-ph.CO astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
We investigate the physical properties and detectability of warm-hot intergalactic medium (WHIM) gas with temperatures in the range $10^5<T<10^7$K around galaxy clusters using simulated galaxy clusters from The Three Hundred project. In simulations with different input physics (GIZMO-SIMBA and Gadget-X), we consistently find that the median gas temperature decreases to the WHIM upper bound, $10^7$K, at $\sim 2 \times R_{200c}$, while the WHIM mass fraction increases with radius until $\sim 3\times R_{200c}$, where it plateaus at $\sim 70$ per cent.By simulating X-ray emission from all gas components, we find that the WHIM contribution to the soft X-ray band (0.2 - 2.3 keV) increases with radius but eventually plateaus at larger distances. The differences between the two simulations become more pronounced at higher redshifts and larger radii. Finally, after accounting for observational effects, primarily by removing (sub)halos, we predict that the signal-to-noise ratio of the X-ray signal obtained by stacking the eRASS1 galaxy cluster catalogue will be $\sim 7$ for GIZMO-SIMBA and $\sim 21$ for Gadget-X.
|
[
{
"version": "v1",
"created": "Thu, 06 Mar 2025 22:26:11 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Li",
"Renjie",
""
],
[
"Cui",
"Weiguang",
""
],
[
"Liu",
"Ang",
""
],
[
"Wang",
"Huiyuan",
""
],
[
"Srivastava",
"Atulit",
""
],
[
"Dave",
"Romeel",
""
],
[
"Pearce",
"Frazer R.",
""
]
] |
2503.05151
|
Akio Kawauchi
|
Akio Kawauchi
|
Revised note on surface-link of trivial components
|
Minor corrections and updated references
| null | null | null |
math.GT
|
http://creativecommons.org/licenses/by/4.0/
|
It is shown that a surface-link of ribbon surface-knot components is a ribbon surface-link if and only if it is a surface-link producing a ribbon surface-link by surgery along a self-trivial 1-handle system. This corrects an earlier statement. This result makes a corrected proof for the claim that every surface-link of trivial surface-knot components with at most one aspheric component is a ribbon surface-link. For non-ribbon surface-links of trivial components with at least two aspheric components constructed in a previous note, it adds the new property that non-ribbonability continues through surgery along any self-trivial 1-handle system.
|
[
{
"version": "v1",
"created": "Fri, 07 Mar 2025 05:19:45 GMT"
},
{
"version": "v2",
"created": "Thu, 10 Jul 2025 07:54:21 GMT"
},
{
"version": "v3",
"created": "Fri, 29 Aug 2025 21:52:01 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Kawauchi",
"Akio",
""
]
] |
2503.05227
|
Jungbae Park
|
Jungbae Park, Heonseok Jang
|
MOHPER: Multi-objective Hyperparameter Optimization Framework for E-commerce Retrieval System
|
Accepted to CIKM '25. This is the author's version of the work. The definitive version of record is published in the Proceedings of the 34th ACM International Conference on Information and Knowledge Management (CIKM '25), https://doi.org/10.1145/3746252.3761496
| null |
10.1145/3746252.3761496
| null |
cs.IR cs.AI
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
E-commerce search optimization has evolved to include a wider range of metrics that reflect user engagement and business objectives. Modern search frameworks now incorporate advanced quality features, such as sales counts and document-query relevance, to better align search results with these goals. Traditional methods typically focus on click-through rate (CTR) as a measure of engagement or relevance, but this can miss true purchase intent, creating a gap between user interest and actual conversions. Joint training with the click-through conversion rate (CTCVR) has become essential for understanding buying behavior, although its sparsity poses challenges for reliable optimization. This study presents MOHPER, a Multi-Objective Hyperparameter Optimization framework for E-commerce Retrieval systems. Utilizing Bayesian optimization and sampling, it jointly optimizes both CTR, CTCVR, and relevant objectives, focusing on engagement and conversion of the users. In addition, to improve the selection of the best configuration from multi-objective optimization, we suggest advanced methods for hyperparameter selection, including a meta-configuration voting strategy and a cumulative training approach that leverages prior optimal configurations, to improve speeds of training and efficiency. Currently deployed in a live setting, our proposed framework substantiates its practical efficacy in achieving a balanced optimization that aligns with both user satisfaction and revenue goals.
|
[
{
"version": "v1",
"created": "Fri, 07 Mar 2025 08:25:08 GMT"
},
{
"version": "v2",
"created": "Thu, 22 May 2025 19:14:29 GMT"
},
{
"version": "v3",
"created": "Sun, 31 Aug 2025 15:44:16 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Park",
"Jungbae",
""
],
[
"Jang",
"Heonseok",
""
]
] |
2503.05320
|
Zitao Fang
|
Zitao Fang, Guodong DU, Shuyang Yu, Yifei Guo, Yiwei Zhang, Yiyao Cao, Jing Li, Ho-Kin Tang, Sim Kuan Goh
|
To See a World in a Spark of Neuron: Disentangling Multi-task Interference for Training-free Model Merging
|
Accepted to EMNLP 2025 Main Conference. This is the camera-ready version. Code: https://ZzzitaoFang.github.io/projects/NeuroMerging/
| null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Fine-tuning pre-trained models on targeted datasets enhances task-specific performance but often comes at the expense of generalization. Model merging techniques, which integrate multiple fine-tuned models into a single multi-task model through task arithmetic, offer a promising solution. However, task interference remains a fundamental challenge, leading to performance degradation and suboptimal merged models. Existing approaches largely overlooked the fundamental roles of neurons, their connectivity, and activation, resulting in a merging process and a merged model that does not consider how neurons relay and process information. In this work, we present the first study that relies on neuronal mechanisms for model merging. Specifically, we decomposed task-specific representations into two complementary neuronal subspaces that regulate input sensitivity and task adaptability. Leveraging this decomposition, we introduced NeuroMerging, a novel merging framework developed to mitigate task interference within neuronal subspaces, enabling training-free model fusion across diverse tasks. Through extensive experiments, we demonstrated that NeuroMerging achieved superior performance compared to existing methods on multi-task benchmarks across both natural language and vision domains. Our findings highlighted the importance of aligning neuronal mechanisms in model merging, offering new insights into mitigating task interference and improving knowledge fusion. Our project is available at https://ZzzitaoFang.github.io/projects/NeuroMerging/.
|
[
{
"version": "v1",
"created": "Fri, 07 Mar 2025 11:00:24 GMT"
},
{
"version": "v2",
"created": "Sun, 25 May 2025 17:17:43 GMT"
},
{
"version": "v3",
"created": "Sun, 31 Aug 2025 04:27:01 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Fang",
"Zitao",
""
],
[
"DU",
"Guodong",
""
],
[
"Yu",
"Shuyang",
""
],
[
"Guo",
"Yifei",
""
],
[
"Zhang",
"Yiwei",
""
],
[
"Cao",
"Yiyao",
""
],
[
"Li",
"Jing",
""
],
[
"Tang",
"Ho-Kin",
""
],
[
"Goh",
"Sim Kuan",
""
]
] |
2503.05403
|
Verena H\"aberle
|
Verena H\"aberle, Xiuqiang He, Linbin Huang, Florian D\"orfler, Steven Low
|
Decentralized Parametric Stability Certificates for Grid-Forming Converter Control
|
13 pages, 15 figures
| null | null | null |
eess.SY cs.SY
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We propose a decentralized framework for guaranteeing the small-signal stability of future power systems with grid-forming converters. Our approach leverages dynamic loop-shifting techniques to compensate for the lack of passivity in the network dynamics and establishes decentralized parametric stability certificates, depending on the local device-level controls and incorporating the effects of the network dynamics. By following practical tuning rules, we are able to ensure plug-and-play operation without centralized coordination. Unlike prior works, our approach accommodates coupled frequency and voltage dynamics, incorporates network dynamics, and does not rely on specific network configurations or operating points, offering a general and scalable solution for the integration of power-electronics-based devices into future power systems. We validate our theoretical stability results through numerical case studies in a high-fidelity simulation model.
|
[
{
"version": "v1",
"created": "Fri, 07 Mar 2025 13:26:55 GMT"
},
{
"version": "v2",
"created": "Thu, 13 Mar 2025 13:51:36 GMT"
},
{
"version": "v3",
"created": "Mon, 17 Mar 2025 14:54:56 GMT"
},
{
"version": "v4",
"created": "Wed, 09 Apr 2025 18:30:07 GMT"
},
{
"version": "v5",
"created": "Tue, 24 Jun 2025 20:07:53 GMT"
},
{
"version": "v6",
"created": "Mon, 01 Sep 2025 18:04:40 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Häberle",
"Verena",
""
],
[
"He",
"Xiuqiang",
""
],
[
"Huang",
"Linbin",
""
],
[
"Dörfler",
"Florian",
""
],
[
"Low",
"Steven",
""
]
] |
2503.05434
|
Michiru Uwabo-Niibo
|
Cristiano Germani, Mohammad Ali Gorji, Michiru Uwabo-Niibo, and Masahide Yamaguchi
|
Peaks sphericity of non-Gaussian random fields
|
37 pages, 4 figures and 1 table. v2 matches the version accepted in JCAP
| null | null | null |
astro-ph.CO gr-qc hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
We formulate the statistics of peaks of non-Gaussian random fields and implement it to study the sphericity of peaks. For non-Gaussianity of the local type, we present a general formalism valid regardless of how large the deviation from Gaussian statistics is. For general types of non-Gaussianity, we provide a framework that applies to any system with a given power spectrum and the corresponding bispectrum in the regime in which contributions from higher-order correlators can be neglected. We present an explicit expression for the most probable values of the sphericity parameters, including the effect of non-Gaussianity on the shape. We show that the effects of small perturbative non-Gaussianity on the sphericity parameters are negligible, as they are even smaller than the subleading Gaussian corrections. In contrast, we find that large non-Gaussianity can significantly distort the peak configurations, making them much less spherical.
|
[
{
"version": "v1",
"created": "Fri, 07 Mar 2025 14:04:56 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 02:05:00 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Germani",
"Cristiano",
""
],
[
"Gorji",
"Mohammad Ali",
""
],
[
"Uwabo-Niibo",
"Michiru",
""
],
[
"Yamaguchi",
"Masahide",
""
]
] |
2503.05530
|
Mathis Randl
|
Shai Bergman, Zhang Ji, Anne-Marie Kermarrec, Diana Petrescu, Rafael Pires, Mathis Randl, Martijn de Vos
|
Leveraging Approximate Caching for Faster Retrieval-Augmented Generation
| null | null |
10.1145/3721146.3721941
| null |
cs.DB cs.LG cs.PF
|
http://creativecommons.org/licenses/by/4.0/
|
Retrieval-augmented generation (RAG) improves the reliability of large language model (LLM) answers by integrating external knowledge. However, RAG increases the end-to-end inference time since looking for relevant documents from large vector databases is computationally expensive. To address this, we introduce Proximity, an approximate key-value cache that optimizes the RAG workflow by leveraging similarities in user queries. Instead of treating each query independently, Proximity reuses previously retrieved documents when similar queries appear, substantially reducing reliance on expensive vector database lookups. To scale efficiently, Proximity employs a locality-sensitive hashing (LSH) scheme that enables fast cache lookups while preserving retrieval accuracy. We evaluate Proximity using the MMLU and MedRAG question answering benchmarks. Our experiments demonstrate that Proximity with our LSH scheme and a realistically skewed MedRAG workload reduces database calls by 78.9% while maintaining database recall and test accuracy. We experiment with different similarity tolerances and cache capacities, and show that the time spent within the Proximity cache remains low and constant (4.8 microseconds) even as the cache grows substantially in size. Our work highlights that approximate caching is a viable and effective strategy for optimizing RAG-based systems.
|
[
{
"version": "v1",
"created": "Fri, 07 Mar 2025 15:54:04 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 13:09:37 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Bergman",
"Shai",
""
],
[
"Ji",
"Zhang",
""
],
[
"Kermarrec",
"Anne-Marie",
""
],
[
"Petrescu",
"Diana",
""
],
[
"Pires",
"Rafael",
""
],
[
"Randl",
"Mathis",
""
],
[
"de Vos",
"Martijn",
""
]
] |
2503.05871
|
Javier Carballo
|
Javier Carballo, Christiana Pantelidou, Benjamin Withers
|
Non-modal effects in black hole perturbation theory: Transient Superradiance
|
30 pages, 13 figures, 1 table. Comments welcome. Added some references. Matches published version
| null |
10.1007/JHEP08(2025)179
| null |
gr-qc hep-th math-ph math.MP
|
http://creativecommons.org/licenses/by/4.0/
|
We study the non-modal stability of black hole spacetimes under linear perturbations. We show that large-amplitude growth can occur at finite time, despite asymptotic decay of linear perturbations. In the example presented, the physical mechanism is a transient form of superradiance, and is qualitatively similar to the transition to turbulence in Navier-Stokes shear flows. As part of the construction we provide a theorem for the positivity of QNM energies, and introduce a truncated-Hamiltonian approach to black hole pseudospectra which does not suffer from convergence issues.
|
[
{
"version": "v1",
"created": "Fri, 07 Mar 2025 19:00:03 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 14:00:35 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Carballo",
"Javier",
""
],
[
"Pantelidou",
"Christiana",
""
],
[
"Withers",
"Benjamin",
""
]
] |
2503.05874
|
Mohammad Chopannavaz
|
Amin Ghodousian, Mohammad Sedigh Chopannavaz, and Witold Pedrycz
|
On the resolution and linear optimization problems subject to a system of bipolar fuzzy relational equalities defined with continuous Archimedean t-norms
| null |
Volume 22, Issue 4 July and August 2025 Pages 137-160
|
10.22111/ijfs.2025.50906.8994
| null |
math.OC
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
This paper considers the linear objective function optimization with respect to a more general class of bipolar fuzzy relational equations, where the fuzzy compositions are defined by an arbitrary continuous Archimedean t-norm. In addition, a faster method for finding a global optimum is proposed. Analytical concepts and properties of the Archimedean bipolar fuzzy equations are investigated and two necessary conditions are presented to conceptualize the feasibility of the problem. It is shown that the feasible solution set can be resulted by a union of the finite number of compact sets, where each compact set is obtained by a function (called admissible function in this paper). Moreover, to accelerate identification of the mentioned compact sets (and therefore, to speed up solution finding), four simplification techniques are presented, which are based on either omitting redundant constraints and/or eliminating unknowns by assigning them a fixed value. Also, three additional simplification techniques are given to reduce the search domain by removing some parts of the feasible region that do not contain optimal solutions. Subsequently, a method is proposed to find an optimal solution for the current linear optimization problems. The proposed method consists of two accelerative strategies that are used during the problem solving process. By the first strategy, the method neglects some candidate solutions that are not optimal, by considering only a subset of admissible functions (called modified functions in this paper). As for the second strategy, a branch-and-bound method is used to delete non-optimal branches. Then, the method is summarized in an algorithm that represents all essential steps of the solution and finally, the whole method is applied in an example that has been chosen in such a way that the various situations are illustrated.
|
[
{
"version": "v1",
"created": "Sat, 08 Feb 2025 15:27:59 GMT"
},
{
"version": "v2",
"created": "Fri, 29 Aug 2025 19:39:31 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Ghodousian",
"Amin",
""
],
[
"Chopannavaz",
"Mohammad Sedigh",
""
],
[
"Pedrycz",
"Witold",
""
]
] |
2503.06397
|
Yanyu Zhu
|
Yanyu Zhu, Lichen Bai, Jintao Xu, Hai-tao Zheng
|
Removing Averaging: Personalized Lip-Sync Driven Characters Based on Identity Adapter
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Recent advances in diffusion-based lip-syncing generative models have demonstrated their ability to produce highly synchronized talking face videos for visual dubbing. Although these models excel at lip synchronization, they often struggle to maintain fine-grained control over facial details in generated images. In this work, we identify "lip averaging" phenomenon where the model fails to preserve subtle facial details when dubbing unseen in-the-wild videos. This issue arises because the commonly used UNet backbone primarily integrates audio features into visual representations in the latent space via cross-attention mechanisms and multi-scale fusion, but it struggles to retain fine-grained lip details in the generated faces. To address this issue, we propose UnAvgLip, which extracts identity embeddings from reference videos to generate highly faithful facial sequences while maintaining accurate lip synchronization. Specifically, our method comprises two primary components: (1) an Identity Perceiver module that encodes facial embeddings to align with conditioned audio features; and (2) an ID-CrossAttn module that injects facial embeddings into the generation process, enhancing model's capability of identity retention. Extensive experiments demonstrate that, at a modest training and inference cost, UnAvgLip effectively mitigates the "averaging" phenomenon in lip inpainting, significantly preserving unique facial characteristics while maintaining precise lip synchronization. Compared with the original approach, our method demonstrates significant improvements of 5% on the identity consistency metric and 2% on the SSIM metric across two benchmark datasets (HDTF and LRW).
|
[
{
"version": "v1",
"created": "Sun, 09 Mar 2025 02:36:31 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 07:24:05 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Zhu",
"Yanyu",
""
],
[
"Bai",
"Lichen",
""
],
[
"Xu",
"Jintao",
""
],
[
"Zheng",
"Hai-tao",
""
]
] |
2503.06645
|
Jiangtao Duan
|
Jiangtao Duan, Jushan Bai, Xu Han
|
Taxonomy and Estimation of Multiple Breakpoints in High-Dimensional Factor Models
| null | null | null | null |
econ.EM
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper investigates the estimation of high-dimensional factor models in which factor loadings undergo an unknown number of structural changes over time. Given that a model with multiple changes in factor loadings can be observationally indistinguishable from one with constant loadings but varying factor variances, this reduces the high-dimensional structural change problem to a lower-dimensional one. Due to the presence of multiple breakpoints, the factor space may expand, potentially causing the pseudo factor covariance matrix within some regimes to be singular. We define two types of breakpoints: {\bf a singular change}, where the number of factors in the combined regime exceeds the minimum number of factors in the two separate regimes, and {\bf a rotational change}, where the number of factors in the combined regime equals that in each separate regime. Under a singular change, we derive the properties of the small eigenvalues and establish the consistency of the QML estimators. Under a rotational change, unlike in the single-breakpoint case, the pseudo factor covariance matrix within each regime can be either full rank or singular, yet the QML estimation error for the breakpoints remains stably bounded. We further propose an information criterion (IC) to estimate the number of breakpoints and show that, with probability approaching one, it accurately identifies the true number of structural changes. Monte Carlo simulations confirm strong finite-sample performance. Finally, we apply our method to the FRED-MD dataset, identifying five structural breaks in factor loadings between 1959 and 2024.
|
[
{
"version": "v1",
"created": "Sun, 09 Mar 2025 14:44:22 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 15:44:17 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Duan",
"Jiangtao",
""
],
[
"Bai",
"Jushan",
""
],
[
"Han",
"Xu",
""
]
] |
2503.06794
|
Yizheng Sun
|
Yizheng Sun, Hao Li, Chang Xu, Hongpeng Zhou, Chenghua Lin, Riza Batista-Navarro, Jingyuan Sun
|
Does Acceleration Cause Hidden Instability in Vision Language Models? Uncovering Instance-Level Divergence Through a Large-Scale Empirical Study
|
Accepted to EMNLP 2025 Main Conference
| null | null | null |
cs.CV cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Vision-Language Models (VLMs) are powerful yet computationally intensive for widespread practical deployments. To address such challenge without costly re-training, post-training acceleration techniques like quantization and token reduction are extensively explored. However, current acceleration evaluations primarily target minimal overall performance degradation, overlooking a crucial question: does the accelerated model still give the same answers to the same questions as it did before acceleration? This is vital for stability-centered industrial applications where consistently correct answers for specific, known situations are paramount, such as in AI-based disease diagnosis. We systematically investigate this for accelerated VLMs, testing four leading models (LLaVA-1.5, LLaVA-Next, Qwen2-VL, Qwen2.5-VL) with eight acceleration methods on ten multi-modal benchmarks. Our findings are stark: despite minimal aggregate performance drops, accelerated models changed original answers up to 20% of the time. Critically, up to 6.5% of these changes converted correct answers to incorrect. Input perturbations magnified these inconsistencies, and the trend is confirmed by case studies with the medical VLM LLaVA-Med. This research reveals a significant oversight in VLM acceleration, stressing an urgent need for instance-level stability checks to ensure trustworthy real-world deployment.
|
[
{
"version": "v1",
"created": "Sun, 09 Mar 2025 22:16:48 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Mar 2025 14:34:14 GMT"
},
{
"version": "v3",
"created": "Tue, 20 May 2025 14:31:45 GMT"
},
{
"version": "v4",
"created": "Sun, 31 Aug 2025 23:37:11 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Sun",
"Yizheng",
""
],
[
"Li",
"Hao",
""
],
[
"Xu",
"Chang",
""
],
[
"Zhou",
"Hongpeng",
""
],
[
"Lin",
"Chenghua",
""
],
[
"Batista-Navarro",
"Riza",
""
],
[
"Sun",
"Jingyuan",
""
]
] |
2503.07169
|
Anders Sundelin
|
Anders Sundelin and Javier Gonzalez-Huerta and Krzysztof Wnuk
|
Reducing Friction in Cloud Migration of Services
|
Submitted to JSS In-Practice track Mars 8, 2025. Revision 1 resubmitted July 29, 2025
| null | null | null |
cs.SE
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Public cloud services are integral to modern software development, offering scalability and flexibility to organizations. Based on customer requests, a large product development organization considered migrating the microservice-based product deployments of a large customer to a public cloud provider.
We conducted an exploratory single-case study, utilizing quantitative and qualitative data analysis to understand how and why deployment costs would change when transitioning the product from a private to a public cloud environment while preserving the software architecture. We also isolated the major factors driving the changes in deployment costs.
We found that switching to the customer-chosen public cloud provider would increase costs by up to 50\%, even when sharing some resources between deployments, and limiting the use of expensive cloud services such as security log analyzers. A large part of the cost was related to the sizing and license costs of the existing relational database, which was running on Virtual Machines in the cloud. We also found that existing system integrators, using the product via its API, were likely to use the product inefficiently, in many cases causing at least 10\% more load to the system than needed.
From a deployment cost perspective, successful migration to a public cloud requires considering the entire system architecture, including services like relational databases, value-added cloud services, and enabled product features. Our study highlights the importance of leveraging end-to-end usage data to assess and manage these cost drivers effectively, especially in environments with elastic costs, such as public cloud deployments.
|
[
{
"version": "v1",
"created": "Mon, 10 Mar 2025 10:47:27 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 13:40:08 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Sundelin",
"Anders",
""
],
[
"Gonzalez-Huerta",
"Javier",
""
],
[
"Wnuk",
"Krzysztof",
""
]
] |
2503.07221
|
Christian P\"otzsche
|
Christian P\"otzsche, Robert Skiba
|
Evans function, parity and nonautonomous bifurcations
| null | null |
10.1017/prm.2025.10062
| null |
math.DS
|
http://creativecommons.org/licenses/by/4.0/
|
The concept of parity due to Fitzpatrick, Pejsachowicz and Rabier is a central tool in the abstract bifurcation theory of nonlinear Fredholm operators. In this paper, we relate the parity to the Evans function, which is widely used in the stability analysis for traveling wave solutions to evolutionary PDEs.
As application we obtain a flexible and general condition yielding local bifurcations of specific bounded entire solutions to (Carath{\'e}odory) differential equations. These bifurcations are intrinsically nonautonomous in the sense that the assumptions implying them cannot be fulfilled for autonomous or periodic temporal forcings. In addition, we demonstrate that Evans functions are strictly related to the dichotomy spectrum and hyperbolicity, which play a crucial role in studying the existence of bounded solutions on the whole real line and therefore the recent field of nonautonomous bifurcation theory. Finally, by means of non-trivial examples we illustrate the applicability of our methods.
|
[
{
"version": "v1",
"created": "Mon, 10 Mar 2025 12:03:48 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Pötzsche",
"Christian",
""
],
[
"Skiba",
"Robert",
""
]
] |
2503.07263
|
Haolin He
|
Haolin He, Ce Zhu, Le Zhang, Yipeng Liu, Xiao Xu, Yuqian Chen, Leo Zekelman, Jarrett Rushmore, Yogesh Rathi, Nikos Makris, Lauren J. O'Donnell, and Fan Zhang
|
DeepNuParc: A Novel Deep Clustering Framework for Fine-scale Parcellation of Brain Nuclei Using Diffusion MRI Tractography
|
Already accepted by NeuroImage
| null | null | null |
eess.IV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Brain nuclei are clusters of anatomically distinct neurons that serve as important hubs for processing and relaying information in various neural circuits. Fine-scale parcellation of the brain nuclei is vital for a comprehensive understanding of its anatomico-functional correlations. Diffusion MRI tractography is an advanced imaging technique that can estimate the brain's white matter structural connectivity to potentially reveal the topography of the nuclei of interest for studying its subdivisions. In this work, we present a deep clustering pipeline, namely DeepNuParc, to perform automated, fine-scale parcellation of brain nuclei using diffusion MRI tractography. First, we incorporate a newly proposed deep learning approach to enable accurate segmentation of the nuclei of interest directly on the dMRI data. Next, we design a novel streamline clustering-based structural connectivity feature for a robust representation of voxels within the nuclei. Finally, we improve the popular joint dimensionality reduction and k-means clustering approach to enable nuclei parcellation at a finer scale. We demonstrate DeepNuParc on two important brain structures, i.e. the amygdala and the thalamus, that are known to have multiple anatomically and functionally distinct nuclei subdivisions. Experimental results show that DeepNuParc enables consistent parcellation of the nuclei into multiple parcels across multiple subjects and achieves good correspondence with the widely used coarse-scale atlases. Our codes are available at https://github.com/HarlandZZC/deep_nuclei_parcellation.
|
[
{
"version": "v1",
"created": "Mon, 10 Mar 2025 12:47:12 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 03:58:07 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"He",
"Haolin",
""
],
[
"Zhu",
"Ce",
""
],
[
"Zhang",
"Le",
""
],
[
"Liu",
"Yipeng",
""
],
[
"Xu",
"Xiao",
""
],
[
"Chen",
"Yuqian",
""
],
[
"Zekelman",
"Leo",
""
],
[
"Rushmore",
"Jarrett",
""
],
[
"Rathi",
"Yogesh",
""
],
[
"Makris",
"Nikos",
""
],
[
"O'Donnell",
"Lauren J.",
""
],
[
"Zhang",
"Fan",
""
]
] |
2503.07353
|
Yaroslava Lochman
|
Carl Olsson, Yaroslava Lochman, Johan Malmport, Christopher Zach
|
Certifiably Optimal Anisotropic Rotation Averaging
| null | null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Rotation averaging is a key subproblem in applications of computer vision and robotics. Many methods for solving this problem exist, and there are also several theoretical results analyzing difficulty and optimality. However, one aspect that most of these have in common is a focus on the isotropic setting, where the intrinsic uncertainties in the measurements are not fully incorporated into the resulting optimization task. Recent empirical results suggest that moving to an anisotropic framework, where these uncertainties are explicitly included, can result in an improvement of solution quality. However, global optimization for rotation averaging has remained a challenge in this scenario. In this work we show how anisotropic costs can be incorporated in certifiably optimal rotation averaging. We also demonstrate how existing solvers, designed for isotropic situations, fail in the anisotropic setting. Finally, we propose a stronger relaxation and empirically show that it recovers global optima in all tested datasets and leads to more accurate reconstructions in almost all scenes.
|
[
{
"version": "v1",
"created": "Mon, 10 Mar 2025 14:09:27 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 14:50:06 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Olsson",
"Carl",
""
],
[
"Lochman",
"Yaroslava",
""
],
[
"Malmport",
"Johan",
""
],
[
"Zach",
"Christopher",
""
]
] |
2503.07898
|
Ahmed Mahmoud
|
Massimiliano Meneghin, Ahmed H. Mahmoud
|
Disaggregated Design for GPU-Based Volumetric Data Structures
|
This preprint has not undergone peer review (when applicable) or any post-submission improvements or corrections. The Version of Record of this contribution is published in Springer's Lecture Notes in Computer Science, Volume 15902 (Euro-Par 2025: Parallel Processing), and is available online at https://doi.org/10.1007/978-3-031-99872-0_11
| null | null | null |
cs.DC cs.PF
|
http://creativecommons.org/licenses/by/4.0/
|
Volumetric data structures typically prioritize data locality, focusing on efficient memory access patterns. This singular focus can neglect other critical performance factors, such as occupancy, communication, and kernel fusion. We introduce a novel \emph{disaggregated} design that rebalances trade-offs between locality and these objectives -- reducing communication overhead on distributed memory architectures, mitigating register pressure in complex boundary conditions, and enabling kernel fusion. We provide a thorough analysis of its benefits on a single-node multi-GPU Lattice Boltzmann Method (LBM) solver. Our evaluation spans dense, block-sparse, and multi-resolution discretizations, demonstrating our design's flexibility and efficiency. Leveraging this approach, we achieve up to a $3\times$ speedup over state-of-the-art solutions.
|
[
{
"version": "v1",
"created": "Mon, 10 Mar 2025 22:39:35 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 15:35:00 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Meneghin",
"Massimiliano",
""
],
[
"Mahmoud",
"Ahmed H.",
""
]
] |
2503.08167
|
Salvatore Capozziello
|
Salvatore Capozziello, Sara Cesare, and Carmen Ferrara
|
Extended Geometric Trinity of Gravity
|
24 pages, 4 figures
|
Eur. Phys. J. C (2025) 85:932
|
10.1140/epjc/s10052-025-14440-2
| null |
gr-qc astro-ph.CO hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
Extensions of equivalent representations of gravity are discussed in the metric-affine framework. First, we focus on: (i) General Relativity, based upon the metric tensor whose dynamics is given by the Ricci curvature scalar $R$; (ii) the Teleparallel Equivalent of General Relativity, based on tetrads and spin connection whose dynamics is given by the torsion scalar $T$; (iii) the Symmetric Teleparallel Equivalent of General Relativity, formulated with respect to both the metric tensor and the affine connection and characterized by the non-metric scalar $Q$ with the role of gravitational field. They represent the so-called Geometric Trinity of Gravity, because, even if based on different frameworks and different dynamical variables, such as curvature, torsion, and non-metricity, they express the same gravitational dynamics. Starting from this framework, we construct their extensions with the aim to study possible equivalence. We discuss the straightforward extension of General Relativity, the $f(R)$ gravity, where $f(R)$ is an arbitrary function of the Ricci scalar. With this paradigm in mind, we take into account $f(T)$ and $f(Q)$ extensions showing that they are not equivalent to $f(R)$. Dynamical equivalence is achieved if boundary terms are considered, that is $f(T-\tilde{B})$ and $f(Q-B)$ theories. The latter are the extensions of Teleparallel Equivalent of General Relativity and Symmetric Teleparallel of General Relativity, respectively. We obtain that $f(R)$, $f(T-\tilde{B})$, and $f(Q-B)$ form the Extended Geometric Trinity of Gravity. The aim is to show that also if dynamics are equivalent, foundations of theories of gravity can be very different.
|
[
{
"version": "v1",
"created": "Tue, 11 Mar 2025 08:30:27 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Capozziello",
"Salvatore",
""
],
[
"Cesare",
"Sara",
""
],
[
"Ferrara",
"Carmen",
""
]
] |
2503.08346
|
Chanyoung Kim
|
Chanyoung Kim, Dayun Ju, Jinyeong Kim, Woojung Han, Roberto Alcover-Couso and Seong Jae Hwang
|
Pathology-Aware Adaptive Watermarking for Text-Driven Medical Image Synthesis
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
As recent text-conditioned diffusion models have enabled the generation of high-quality images, concerns over their potential misuse have also grown. This issue is critical in the medical domain, where text-conditioned generated medical images could enable insurance fraud or falsified records, highlighting the urgent need for reliable safeguards against unethical use. While watermarking techniques have emerged as a promising solution in general image domains, their direct application to medical imaging presents significant challenges. A key challenge is preserving fine-grained disease manifestations, as even minor distortions from a watermark may lead to clinical misinterpretation, which compromises diagnostic integrity. To overcome this gap, we present MedSign, a deep learning-based watermarking framework specifically designed for text-to-medical image synthesis, which preserves pathologically significant regions by adaptively adjusting watermark strength. Specifically, we generate a pathology localization map using cross-attention between medical text tokens and the diffusion denoising network, aggregating token-wise attention across layers, heads, and time steps. Leveraging this map, we optimize the LDM decoder to incorporate watermarking during image synthesis, ensuring cohesive integration while minimizing interference in diagnostically critical regions. Experimental results show that our MedSign preserves diagnostic integrity while ensuring watermark robustness, achieving state-of-the-art performance in image quality and detection accuracy on MIMIC-CXR and OIA-ODIR datasets.
|
[
{
"version": "v1",
"created": "Tue, 11 Mar 2025 11:55:14 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 01:30:25 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Kim",
"Chanyoung",
""
],
[
"Ju",
"Dayun",
""
],
[
"Kim",
"Jinyeong",
""
],
[
"Han",
"Woojung",
""
],
[
"Alcover-Couso",
"Roberto",
""
],
[
"Hwang",
"Seong Jae",
""
]
] |
2503.08402
|
Umananda Dev Goswami
|
Gayatri Mohan, Ronit Karmakar, Rupam Jyoti Borah and Umananda Dev Goswami
|
Strong Lensing Effect and Quasinormal Modes of Oscillations of Black Holes in $\boldsymbol{f(R,T)}$ Gravity Theory
|
27 pages, 19 figures; Published version
|
Physics of the Dark Universe 49 (2025) 102007
|
10.1016/j.dark.2025.102007
| null |
gr-qc
|
http://creativecommons.org/licenses/by/4.0/
|
In this work, we analyze the strong lensing phenomenon and quasinormal modes (QNMs) in the case of black holes (BHs) surrounded by fluids within the framework of $f(R,T)$ gravity, adopting a minimally coupled model of the theory. Our analysis is conducted for three surrounding fields corresponding to three different values of the parameter $\omega$ of the equations of state, each representing a unique class of BH solutions. A universal method developed by V.~Bozza is employed for strong lensing analysis and the WKB approximation method to compute the QNMs of oscillation of the BHs. The influences of the model parameters $\beta$ and $c_2$ on the deflection angle and associated lensing coefficients are analyzed. Our findings on lensing reveal that smaller values of $\beta$ and $c_2$ cause photon divergence at larger impact parameters as well as the lensing results converge to the Schwarzschild limit. Extending the analysis to the supermassive BH Sgr A*, we examine the outermost Einstein rings, estimate three lensing observables: angular position $\vartheta_{\infty}$, angular separation $s$ and relative magnification $r_\text{mag}$ for the BHs. For a specific values of $\beta$ and $c_2$, BHs with different field configurations exhibit substantial variations in their observable properties. The variation of amplitude and damping of QNMs with respect to the model parameter $\beta$ and $c_2$ is analyzed for the BHs. We found that the $\beta$ parameter has a direct correlation with the amplitude and an inverse relation with the damping of the QNMs, while $c_2$ has direct correlation with amplitude as well as damping. Further, we use the time domain analysis to verify the results and found a good match between the two methods.
|
[
{
"version": "v1",
"created": "Tue, 11 Mar 2025 13:06:11 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 11:30:09 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Mohan",
"Gayatri",
""
],
[
"Karmakar",
"Ronit",
""
],
[
"Borah",
"Rupam Jyoti",
""
],
[
"Goswami",
"Umananda Dev",
""
]
] |
2503.08447
|
Xiaoxuan Chu
|
Kiera Cassar, Zhen Wang, Xiaoxuan Chu, Elke-Caroline Aschenauer
|
Investigating the broadening phenomenon in two-particle correlations induced by gluon saturation
|
5 pages, 3 figures
| null |
10.1103/7jxl-8pzn
| null |
hep-ph nucl-ex nucl-th
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
It has been found that the gluon density inside the proton grows rapidly at small momentum fractions. Quantum Chromodynamics (QCD) predicts that this growth can be regulated by nonlinear effects, ultimately leading to gluon saturation. Within the color glass condensate framework, nonlinear QCD effects are predicted to suppress and broaden back-to-back angular correlations in collisions involving heavy nuclei. While suppression has been observed in various experiments in $d/p$$+$A collisions compared to $p$$+$$p$ collisions, the predicted broadening remains unobserved. This study investigates the contributions of intrinsic transverse momentum ($k_T$), which is associated with saturation physics, as well as parton showers and transverse motion from fragmentation ($p_T^{\mathrm{frag}}$), which are not saturation dependent, to the width of the correlation function. Our findings show that the non-saturation dependent effects, especially the initial-state parton shower and $p_T^{\mathrm{frag}}$, which occur independently of the collision system, smear the back-to-back correlation more than gluon saturation does, making the broadening phenomenon difficult to observe.
|
[
{
"version": "v1",
"created": "Tue, 11 Mar 2025 13:56:38 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Cassar",
"Kiera",
""
],
[
"Wang",
"Zhen",
""
],
[
"Chu",
"Xiaoxuan",
""
],
[
"Aschenauer",
"Elke-Caroline",
""
]
] |
2503.08504
|
Cheng Zhang
|
Xing Wang, An Zhang and Cheng Zhang
|
Strichartz estimates for orthonormal systems on compact manifolds
|
28 pages, 3 figures
| null | null | null |
math.AP math-ph math.CA math.MP math.SP
|
http://creativecommons.org/licenses/by/4.0/
|
We establish new Strichartz estimates for orthonormal systems on compact Riemannian manifolds in the case of wave, Klein-Gordon and fractional Schr\"odinger equations. Our results generalize the classical (single-function) Strichartz estimates on compact manifolds by Kapitanski, Burq-G\'erard-Tzvetkov, Dinh, and extend the Euclidean orthonormal version by Frank-Lewin-Lieb-Seiringer, Frank-Sabin, Bez-Lee-Nakamura. On the flat torus, our new results for the Schr\"odinger equation cover prior work of Nakamura, which exploits the dispersive estimate of Kenig-Ponce-Vega. We achieve sharp results on compact manifolds by combining the frequency localized dispersive estimates for small time intervals with the duality principle due to Frank-Sabin. We construct examples to show these results can be saturated on the sphere, and we can improve them on the flat torus by using Bourgain-Demeter's decoupling theorem to obtain new decoupling inequalities for certain non-smooth hypersurfaces. As an application, we obtain the well-posedness of infinite systems of dispersive equations with Hartree-type nonlinearity.
|
[
{
"version": "v1",
"created": "Tue, 11 Mar 2025 14:55:47 GMT"
},
{
"version": "v2",
"created": "Wed, 09 Apr 2025 09:35:04 GMT"
},
{
"version": "v3",
"created": "Thu, 10 Apr 2025 13:36:38 GMT"
},
{
"version": "v4",
"created": "Sat, 30 Aug 2025 15:32:27 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Wang",
"Xing",
""
],
[
"Zhang",
"An",
""
],
[
"Zhang",
"Cheng",
""
]
] |
2503.08628
|
Satrio Adi Rukmono
|
Satrio Adi Rukmono, Filip Zamfirov, Lina Ochoa, Floris Pex, Michel Chaudron
|
An Explanation of Software Architecture Explanations
| null | null | null | null |
cs.SE
|
http://creativecommons.org/licenses/by/4.0/
|
Software architecture knowledge transfer is essential for software development, but related documentation is often incomplete or ambiguous, making oral explanations a common means. Our broader aim is to explore how such explanations might be better supported and eventually automated; as a prerequisite, we first investigate how explanations are actually conducted in practice across five areas: explanation topics, explanation plans, supporting artefacts, typical questions, and expectations and challenges. We report on semi-structured interviews with 17 software professionals across diverse organisations and countries. Our findings include that explanations must balance problem- and technical-domain while considering the explainee's role, experience, and goals. Moreover, explainees frequently seek not only structure and behaviour, but also decision rationale. We propose the Explanation Window, a framework for focusing information by adjusting functionality scope and detail, and emphasise the importance of including system context. These findings provide an empirical basis for improving architecture explanations and guiding future work on tool support and automation.
|
[
{
"version": "v1",
"created": "Tue, 11 Mar 2025 17:16:03 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 15:00:26 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Rukmono",
"Satrio Adi",
""
],
[
"Zamfirov",
"Filip",
""
],
[
"Ochoa",
"Lina",
""
],
[
"Pex",
"Floris",
""
],
[
"Chaudron",
"Michel",
""
]
] |
2503.08657
|
Jackson Fliss
|
Jackson R. Fliss
|
Massive fields and Wilson spools in JT gravity
|
25 pages + appendices; one sphere, one disc, one trumpet, one cone, one football
|
SciPost Phys. 19, 065 (2025)
|
10.21468/SciPostPhys.19.3.065
| null |
hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
We give a prescription for minimally coupling massive matter to JT gravity with either sign of cosmological constant directly in its formulation as a topological BF theory. This coupling takes the form of a `Wilson spool,' originally introduced in the context of three-dimensional gravity. The Wilson spool expresses the exact one-loop partition function as the integral over a Wilson loop operator. We construct the spool by considering the partition function of a massive scalar field on Euclidean dS$_2$ and on Euclidean AdS$_2$. We discuss its extension to other geometries (including the `trumpet' and conical defect geometries) and its relation to the three-dimensional spool through dimensional reduction.
|
[
{
"version": "v1",
"created": "Tue, 11 Mar 2025 17:47:02 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Fliss",
"Jackson R.",
""
]
] |
2503.08901
|
Silvia Neri
|
Silvia Neri, Walter Metzner, and Dirk Manske
|
Collective mode spectroscopy in time-reversal symmetry breaking superconductors
| null | null | null | null |
cond-mat.supr-con
|
http://creativecommons.org/licenses/by/4.0/
|
Time-reversal symmetry breaking (TRSB) superconductors show a rich collective mode spectrum. In general, collective excitations in superconductors can provide crucial information on the symmetry of the broken phase, in particular, serving as a fingerprint for determining the groundstate gap symmetry. In this work, we consider several even parity two-dimensional TRSB superconductors characterized by an order parameter of the form $\Delta = \Delta_1 + i\Delta_2$. We provide a classification scheme of the collective excitations in the above systems as a function of the ratio between the components $\Delta_1/\Delta_2$. In order to excite the modes in the systems we have adopted two different probes: a quench of the condensate symmetry and a finite momentum transfer induced by an external electric field. Both methods allow us to excite and characterize the different modes in the spectra. To further interpret the results of the numerical calculations we provide a Ginzburg-Landau analysis and we construct a dynamical theory, deriving the linearized equations of motion in the pseudospin formalism. Our results could help distinguish between different order parameters symmetries of a TRSB superconducting condensate and estimate the magnitude of its different components.
|
[
{
"version": "v1",
"created": "Tue, 11 Mar 2025 21:27:26 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 13:49:09 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Neri",
"Silvia",
""
],
[
"Metzner",
"Walter",
""
],
[
"Manske",
"Dirk",
""
]
] |
2503.09146
|
Linli Yao
|
Linli Yao, Haoning Wu, Kun Ouyang, Yuanxing Zhang, Caiming Xiong, Bei Chen, Xu Sun, Junnan Li
|
Generative Frame Sampler for Long Video Understanding
|
ACL 2025 Findings. Code: https://github.com/yaolinli/GenS
| null | null | null |
cs.CV cs.MM
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Despite recent advances in Video Large Language Models (VideoLLMs), effectively understanding long-form videos remains a significant challenge. Perceiving lengthy videos containing thousands of frames poses substantial computational burden. To mitigate this issue, this paper introduces Generative Frame Sampler (GenS), a plug-and-play module integrated with VideoLLMs to facilitate efficient lengthy video perception. Built upon a lightweight VideoLLM, GenS leverages its inherent vision-language capabilities to identify question-relevant frames. To facilitate effective retrieval, we construct GenS-Video-150K, a large-scale video instruction dataset with dense frame relevance annotations. Extensive experiments demonstrate that GenS consistently boosts the performance of various VideoLLMs, including open-source models (Qwen2-VL-7B, Aria-25B, VILA-40B, LLaVA-Video-7B/72B) and proprietary assistants (GPT-4o, Gemini). When equipped with GenS, open-source VideoLLMs achieve impressive state-of-the-art results on long-form video benchmarks: LLaVA-Video-72B reaches 66.8 (+4.3) on LongVideoBench and 77.0 (+2.7) on MLVU, while Aria obtains 39.2 on HourVideo surpassing the Gemini-1.5-pro by 1.9 points. We will release all datasets and models at https://generative-sampler.github.io.
|
[
{
"version": "v1",
"created": "Wed, 12 Mar 2025 08:16:39 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 09:52:40 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Yao",
"Linli",
""
],
[
"Wu",
"Haoning",
""
],
[
"Ouyang",
"Kun",
""
],
[
"Zhang",
"Yuanxing",
""
],
[
"Xiong",
"Caiming",
""
],
[
"Chen",
"Bei",
""
],
[
"Sun",
"Xu",
""
],
[
"Li",
"Junnan",
""
]
] |
2503.09675
|
Shangwen Zhu
|
Shangwen Zhu, Han Zhang, Zhantao Yang, Qianyu Peng, Zhao Pu, Huangji Wang, Fan Cheng
|
Accelerating Diffusion Sampling via Exploiting Local Transition Coherence
| null | null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Text-based diffusion models have made significant breakthroughs in generating high-quality images and videos from textual descriptions. However, the lengthy sampling time of the denoising process remains a significant bottleneck in practical applications. Previous methods either ignore the statistical relationships between adjacent steps or rely on attention or feature similarity between them, which often only works with specific network structures. To address this issue, we discover a new statistical relationship in the transition operator between adjacent steps, focusing on the relationship of the outputs from the network. This relationship does not impose any requirements on the network structure. Based on this observation, we propose a novel training-free acceleration method called LTC-Accel, which uses the identified relationship to estimate the current transition operator based on adjacent steps. Due to no specific assumptions regarding the network structure, LTC-Accel is applicable to almost all diffusion-based methods and orthogonal to almost all existing acceleration techniques, making it easy to combine with them. Experimental results demonstrate that LTC-Accel significantly speeds up sampling in text-to-image and text-to-video synthesis while maintaining competitive sample quality. Specifically, LTC-Accel achieves a speedup of 1.67-fold in Stable Diffusion v2 and a speedup of 1.55-fold in video generation models. When combined with distillation models, LTC-Accel achieves a remarkable 10-fold speedup in video generation, allowing real-time generation of more than 16FPS.
|
[
{
"version": "v1",
"created": "Wed, 12 Mar 2025 17:52:09 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 08:14:03 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Zhu",
"Shangwen",
""
],
[
"Zhang",
"Han",
""
],
[
"Yang",
"Zhantao",
""
],
[
"Peng",
"Qianyu",
""
],
[
"Pu",
"Zhao",
""
],
[
"Wang",
"Huangji",
""
],
[
"Cheng",
"Fan",
""
]
] |
2503.09683
|
Ben Jaderberg
|
Ben Jaderberg, George Pennington, Kate V. Marshall, Lewis W. Anderson, Abhishek Agarwal, Lachlan P. Lindoy, Ivan Rungger, Stefano Mensa, Jason Crain
|
Variational preparation of normal matrix product states on quantum computers
|
19 pages, 8 figures
| null | null | null |
quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Preparing matrix product states (MPSs) on quantum computers is an essential routine in the simulation of many-body physics. However, widely-used schemes based on staircase circuits are often too deep to execute on current hardware. Here we demonstrate that MPSs with short-range correlations can be prepared with shallow circuits by leveraging heuristics from approximate quantum compiling (AQC). We achieve this with ADAPT-AQC, an adaptive-ansatz preparation algorithm, and introduce a generalised initialisation procedure for the existing AQC-Tensor algorithm. We first compare these methods for the task of preparing a molecular electronic structure ground state. We then use them to prepare an antiferromagnetic (AFM) ground state of the 50-site Heisenberg XXZ spin chain near the AFM-XY phase boundary. Through the execution of circuits with up to 59 CZ depth and 1251 CZ gates, we perform a global quench and observe the relaxation of magnetic ordering in a parameter regime previously inaccessible due to deep ground state preparation circuits. Our results demonstrate how the integration of quantum and classical resources can push the boundary of what can be studied on quantum computers.
|
[
{
"version": "v1",
"created": "Wed, 12 Mar 2025 18:00:00 GMT"
},
{
"version": "v2",
"created": "Fri, 28 Mar 2025 16:29:18 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 09:32:37 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Jaderberg",
"Ben",
""
],
[
"Pennington",
"George",
""
],
[
"Marshall",
"Kate V.",
""
],
[
"Anderson",
"Lewis W.",
""
],
[
"Agarwal",
"Abhishek",
""
],
[
"Lindoy",
"Lachlan P.",
""
],
[
"Rungger",
"Ivan",
""
],
[
"Mensa",
"Stefano",
""
],
[
"Crain",
"Jason",
""
]
] |
2503.09779
|
Lingxiao Zhang
|
Vasileios Chousionis, Sean Li, Lingxiao Zhang
|
Singular integrals on $C^{1,\alpha}$ intrinsic graphs in step 2 Carnot groups
| null | null | null | null |
math.CA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study singular integral operators induced by Calder\'on-Zygmund kernels in any step-$2$ Carnot group $\mathbb{G}$. We show that if such an operator satisfies some natural cancellation conditions then it is $L^2$ bounded on all intrinsic graphs of $C^{1,\alpha}$ functions over vertical hyperplanes that do not have rapid growth at $\infty$.
In particular, the result applies to the Riesz operator $\mathcal{R}$ induced by the kernel $$ \mathsf{R}(z)= \nabla_{\mathbb{G}} \Gamma(z), \quad z\in \mathbb{G}\backslash \{0\}, $$ the horizontal gradient of the fundamental solution of the sub-Laplacian. The $L^2$ boundedness of $\mathcal{R}$ is connected with the question of removability for Lipschitz harmonic functions. As a corollary of our result, we infer that closed subsets of the intrinsic graphs mentioned above are non-removable.
|
[
{
"version": "v1",
"created": "Wed, 12 Mar 2025 19:30:26 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 16:15:46 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Chousionis",
"Vasileios",
""
],
[
"Li",
"Sean",
""
],
[
"Zhang",
"Lingxiao",
""
]
] |
2503.09796
|
Farrukh A. Chishtie
|
Farrukh A. Chishtie
|
Restricting One-Loop radiative effects in quantum gravity: Demonstrating 4D GR as an EFT and its consistent unification with the Standard Model
|
16 pages, LaTeX, further proofs added, accepted at Canadian Journal of Physics
| null | null | null |
hep-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In ``On restricting to one-loop order the radiative effects in quantum gravity" (Brandt, Frenkel, and McKeon, 2020) \cite{Brandt2020}, a Lagrange multiplier (LM) field is introduced into the Einstein-Hilbert action, removing all multi-loop graviton diagrams and confining quantum-gravity corrections to just one loop. The resulting one-loop effective action carries a term proportional to {$\ln(\mu/\Lambda)$}, which they suggest could be experimentally determined, hinting at direct measurements of quantum-gravity effects. We {demonstrate}, however, that {$\mu$ and $\Lambda$} emerge from a chosen renormalization scheme, not from physical observables, implying that {$\ln(\mu/\Lambda)$} signals a finite UV cutoff in this ``LM renormalization scheme.'' Although Newton's constant remains fixed (no running of {$G_N$}), the resulting logarithmic dependence encodes a limited domain of validity for General Relativity (GR) in four dimensions, thereby demonstrating explicitly that 4D GR behaves as an effective field theory (EFT) for energies below the cutoff. {Using the Appelquist-Carazzone decoupling theorem, we prove mathematically that this framework has a well-defined low-energy limit.} We then illustrate how this truncated, renormalized gravity sector can be consistently unified with the Standard Model (SM), yielding a finite and renormalized EFT encompassing both gravity and particle physics up to a scale {$\Lambda_{\text{grav}}$}. Our work represents a breakthrough in theoretical physics with a first successful unification of gravity with the Standard Model through a fully renormalizable and EFT framework.
|
[
{
"version": "v1",
"created": "Wed, 12 Mar 2025 20:01:18 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Mar 2025 04:33:34 GMT"
},
{
"version": "v3",
"created": "Tue, 02 Sep 2025 14:17:51 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Chishtie",
"Farrukh A.",
""
]
] |
2503.09932
|
Vladimir Rosenhaus
|
Xu-Yao Hu and Vladimir Rosenhaus
|
Beyond the Boltzmann equation for weakly coupled quantum fields
|
19 pages, v2: typos fixed, references added
| null | null | null |
hep-th cond-mat.stat-mech hep-ph physics.flu-dyn
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study the kinetic theory of a weakly interacting quantum field. Assuming a state that is close to homogeneous and stationary, we derive a closed kinetic equation for the rate of change of the occupation numbers, perturbatively in the coupling. For a dilute gas, this reproduces the quantum Boltzmann equation, which only accounts for two-to-two scattering processes. Our expression goes beyond this, with terms accounting for multi-particle scattering processes, which are higher order in the density.
|
[
{
"version": "v1",
"created": "Thu, 13 Mar 2025 01:09:49 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 01:37:07 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Hu",
"Xu-Yao",
""
],
[
"Rosenhaus",
"Vladimir",
""
]
] |
2503.10180
|
Rui Li
|
Rui Li, Nicola R. Napolitano, Giuseppe D Ago, Vyacheslav N. Shalyapin, Kai Zhu, Xiaotong Guo, Ran Li, Leon V. E. Koopmans, Chiara Spiniello, Crescenzo Tortora, Francesco La Barbera, Haicheng Feng, Liang Gao, Zhiqi Huang, Koen Kuijken, Hui Li, Linghua Xie, Mario Radovich, Alexey Sergeyev
|
Optical+NIR analysis of a Newly Confirmed Einstein ring at z$\sim$1 from the Kilo-Degree Survey: Dark matter fraction, total and dark matter density slope and IMF
|
14 pages, 4 figures, 2 tables, published by APJL
| null |
10.3847/2041-8213/ade680
| null |
astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
We report the spectroscopic confirmation of a bright blue Einstein ring in the Kilo Degree Survey (KiDS) footprint: the Einstein ``blue eye''. Spectroscopic data from X-Shooter at the Very Large Telescope (VLT) show that the lens is a typical early-type galaxy (ETG) at $z_l=0.9906$, while the background source is a Ly$\alpha$ emitter at $z_s=2.823$. The reference lens modeling was performed on a high-resolution $Y-$band adaptive-optics image from HAWK-I at VLT. Assuming a singular isothermal ellipsoid (SIE) total mass density profile, we inferred an Einstein radius $R_{Ein}=10.47 \pm 0.06$ kpc. The average slope of the total mass density inside the Einstein radius, as determined by a joint analysis of lensing and isotropic Jeans equations is $\gamma_{tot}=2.14^{+0.06}_{-0.07}$, showing no systematic deviation from the slopes of lower redshift galaxies, This can be the evidence of ETGs developing through dry mergers plus moderate dissipationless accretion. Stellar population analysis with 8-band ($gri$ZYJHK$s$) photometries from KiDS and VIKING shows that the total stellar mass of the lens is $M*=(3.95\pm 0.35)\times 10^{11} M_\odot$ (Salpeter Initial Mass Function, IMF), implying a dark matter fraction inside the effective radius to be $f_{\rm DM}=0.307\pm 0.151$. We finally explored the dark matter halo slope and found a strong degeneracy with the dynamic stellar mass. Dark matter adiabatic contraction is needed to explain the posterior distribution of the slope unless IMF heavier than Salpeter is assumed.
|
[
{
"version": "v1",
"created": "Thu, 13 Mar 2025 09:08:24 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 07:27:09 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Li",
"Rui",
""
],
[
"Napolitano",
"Nicola R.",
""
],
[
"Ago",
"Giuseppe D",
""
],
[
"Shalyapin",
"Vyacheslav N.",
""
],
[
"Zhu",
"Kai",
""
],
[
"Guo",
"Xiaotong",
""
],
[
"Li",
"Ran",
""
],
[
"Koopmans",
"Leon V. E.",
""
],
[
"Spiniello",
"Chiara",
""
],
[
"Tortora",
"Crescenzo",
""
],
[
"La Barbera",
"Francesco",
""
],
[
"Feng",
"Haicheng",
""
],
[
"Gao",
"Liang",
""
],
[
"Huang",
"Zhiqi",
""
],
[
"Kuijken",
"Koen",
""
],
[
"Li",
"Hui",
""
],
[
"Xie",
"Linghua",
""
],
[
"Radovich",
"Mario",
""
],
[
"Sergeyev",
"Alexey",
""
]
] |
2503.10201
|
Jolanta Marzec-Ballesteros
|
Thanasis Bouganis and Jolanta Marzec-Ballesteros
|
On an analogue of the doubling method in coding theory
|
27 pages; changes: extended abstract and introduction, corrected typos
| null | null | null |
math.NT cs.IT math.IT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
It is well known that there is a deep relationship between codes and lattices. Concepts from coding theory are related to concepts of lattice theory as, for example, weight enumerators to theta series, MacWilliams identity to Jacobi identity, and Gleason's theorem to Hecke's theorem. In this framework, higher-genus (or multiple) weight enumerators are related to Siegel theta series, which opens up the possibility of introducing concepts from the theory of higher-rank modular forms to coding theory. There has been important work in this direction, for example Runge introduced a coding theory analogue of Siegel's $\Phi$-operator and Nebe analogues of Hecke operators. In this paper, we show that the celebrated Doubling Method from the theory of higher-rank modular forms has a coding theory analogue. Given the impact that the Doubling Method has had in the study of higher-rank modular forms, one may expect that its analogue may prove useful to the study of higher-genus weight enumerators. In this paper we use it to solve an analogue of the "basis problem". That is, we express "cuspidal" polynomials which are invariant under a Clifford-Weil type group as an explicit linear combination of higher-genus weight enumerators of self-dual codes of that type.
|
[
{
"version": "v1",
"created": "Thu, 13 Mar 2025 09:39:19 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 05:32:40 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Bouganis",
"Thanasis",
""
],
[
"Marzec-Ballesteros",
"Jolanta",
""
]
] |
2503.10680
|
Michal Pavelka
|
Nadine Suzan Cetin, Michal Pavelka, Emil Varga
|
A geometric one-fluid model of superfluid helium-4
| null | null | null | null |
cond-mat.supr-con cond-mat.quant-gas cond-mat.stat-mech
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
A standard description of superfluid helium-4 is based on the concept of two components (superfluid and normal), which leads to the so called two-fluid models. However, as there are no two kinds of atoms in helium-4, the two components can not be separated. Superfluid helium-4 is not a mixture of two components, being rather a single fluid with two motions. Here, we present a geometric one-fluid model of superfluid helium-4, which is based on the Hamiltonian formulation of fluid mechanics. The model is derived from the kinetic theory of excitations (treated as an ideal Bose gas under the temperature $1.3K$) and average particle motions. It can be simplified to the Hall-Vinen-Bekharevich-Khalatnikov (HVBK) two-fluid model, where it removes one fitting parameter from the HVBK model, but it also gives extra terms beyond the HVBK model. Actually, we show that the two-fluid models are problematic in case of higher counter-flow velocities, where the usual splitting of total momentum to the superfluid and normal component becomes impossible. Finally, we show how vortex line density may be added to the state variables. The one-fluid model can be seen as a generalization of the two-fluid models that is geometrically consistent, fully compressible, with non-zero superfluid vorticity, and compatible with classical experiments.
|
[
{
"version": "v1",
"created": "Tue, 11 Mar 2025 14:40:12 GMT"
},
{
"version": "v2",
"created": "Fri, 11 Apr 2025 09:36:23 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 14:37:21 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Cetin",
"Nadine Suzan",
""
],
[
"Pavelka",
"Michal",
""
],
[
"Varga",
"Emil",
""
]
] |
2503.10684
|
Zihao Wang
|
Jingwen Deng, Zihao Wang, Shaofei Cai, Anji Liu, Yitao Liang
|
Open-World Skill Discovery from Unsegmented Demonstrations
|
Accepted by ICCV 2025
| null | null | null |
cs.CV cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Learning skills in open-world environments is essential for developing agents capable of handling a variety of tasks by combining basic skills. Online demonstration videos are typically long but unsegmented, making them difficult to segment and label with skill identifiers. Unlike existing methods that rely on sequence sampling or human labeling, we have developed a self-supervised learning-based approach to segment these long videos into a series of semantic-aware and skill-consistent segments. Drawing inspiration from human cognitive event segmentation theory, we introduce Skill Boundary Detection (SBD), an annotation-free temporal video segmentation algorithm. SBD detects skill boundaries in a video by leveraging prediction errors from a pretrained unconditional action-prediction model. This approach is based on the assumption that a significant increase in prediction error indicates a shift in the skill being executed. We evaluated our method in Minecraft, a rich open-world simulator with extensive gameplay videos available online. Our SBD-generated segments improved the average performance of conditioned policies by 63.7% and 52.1% on short-term atomic skill tasks, and their corresponding hierarchical agents by 11.3% and 20.8% on long-horizon tasks. Our method can leverage the diverse YouTube videos to train instruction-following agents. The project page can be found in https://craftjarvis.github.io/SkillDiscovery.
|
[
{
"version": "v1",
"created": "Tue, 11 Mar 2025 18:51:40 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 16:12:04 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Deng",
"Jingwen",
""
],
[
"Wang",
"Zihao",
""
],
[
"Cai",
"Shaofei",
""
],
[
"Liu",
"Anji",
""
],
[
"Liang",
"Yitao",
""
]
] |
2503.10840
|
Yuhao Zhang
|
Yuhao Zhang and Xiangru Xu
|
Efficient Reachability Analysis for Convolutional Neural Networks Using Hybrid Zonotopes
|
Accepted by 2025 American Control Conference (ACC). 8 pages, 1 figure
| null |
10.23919/ACC63710.2025.11107585
| null |
math.OC cs.LG cs.SY eess.SY
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Feedforward neural networks are widely used in autonomous systems, particularly for control and perception tasks within the system loop. However, their vulnerability to adversarial attacks necessitates formal verification before deployment in safety-critical applications. Existing set propagation-based reachability analysis methods for feedforward neural networks often struggle to achieve both scalability and accuracy. This work presents a novel set-based approach for computing the reachable sets of convolutional neural networks. The proposed method leverages a hybrid zonotope representation and an efficient neural network reduction technique, providing a flexible trade-off between computational complexity and approximation accuracy. Numerical examples are presented to demonstrate the effectiveness of the proposed approach.
|
[
{
"version": "v1",
"created": "Thu, 13 Mar 2025 19:45:26 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Zhang",
"Yuhao",
""
],
[
"Xu",
"Xiangru",
""
]
] |
2503.11098
|
Zeliang Wu
|
Zeliang Wu, Jinxian Guo, Zhifei Yu, Wenfeng Huang, Chun-Hua Yuan, Weiping Zhang and L.Q. Chen
|
AI-assisted hyper-dimensional broadband quantum memory with efficiency above 90% in warm atoms
|
7 pages, 4 figures
| null |
10.1038/s41534-025-01092-z
| null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
High-dimensional broadband quantum memory significantly expands quantum information processing capabilities, but the memory efficiency becomes insufficient when extended to high dimensions. We demonstrate an efficient quantum memory for hyper-dimensional photons encoded with orbital angular momentum (OAM) and spin angular momentum (SAM). OAM information is encoded from -5 to +5, combined with SAM encoding, enabling up to 22 dimensions. To ensure high memory efficiency, an artificial intelligence algorithm, a modified Differential Evolution (DE) algorithm using Chebyshev sampling, is developed to obtain a perfect signal-control waveform matching. Memory efficiency is experimentally achieved at 92% for single-mode Gaussian signal, 91% for information dimension of 6 and 80% for dimensional number to 22. The fidelity is achieved up to 99% for single-mode Gaussian signal, 95.5% for OAM information, 97.4% for SAM information, and 92% for whole hyper-dimensional signal, which is far beyond no-cloning limitation. Our results demonstrate superior performance and potential applications in high-dimensional quantum information processing. This achievement provides a crucial foundation for future quantum communication and quantum computing.
|
[
{
"version": "v1",
"created": "Fri, 14 Mar 2025 05:40:22 GMT"
},
{
"version": "v2",
"created": "Thu, 10 Apr 2025 12:02:40 GMT"
},
{
"version": "v3",
"created": "Tue, 02 Sep 2025 09:29:30 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Wu",
"Zeliang",
""
],
[
"Guo",
"Jinxian",
""
],
[
"Yu",
"Zhifei",
""
],
[
"Huang",
"Wenfeng",
""
],
[
"Yuan",
"Chun-Hua",
""
],
[
"Zhang",
"Weiping",
""
],
[
"Chen",
"L. Q.",
""
]
] |
2503.11134
|
Manato Sakai
|
Manato Sakai, Yasuhiro Yamaguchi
|
Analysis of bound and resonant states of doubly heavy tetraquarks with spin $J\leq 2$
|
25 pages, 4 figures, 4 tables
|
Phys.Rev.D 112, 034038 (2025)
|
10.1103/PhysRevD.109.054016
| null |
hep-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Recently, a number of exotic hadrons have been reported in the experiments, and most of these states lie slightly below the threshold. Therefore, these states are considered to be hadronic molecules composed of mesons or baryons. In 2022, the doubly charmed tetraquark $T_{cc}$ was reported by the LHCb experiment, which is considered to be composed of two heavy quarks and two light antiquarks, and located slightly below the $DD^\ast$ threshold. The observation motivates us to study the bound and resonant states of the doubly heavy tetraquarks as two meson systems. We employ the one boson exchange potential as an interaction between two mesons, where one free parameter has been determined to reproduce $T_{cc}$ reported by the LHCb experiment. In our previous work, bound states of $D^{(\ast)}D^{(\ast)}$ and $B^{(\ast)}B^{(\ast)}$ molecules were studied, where we respected the heavy quark spin symmetry (HQS). Using the same framework, we study not only bound but also resonant states with $J\leq 2$ in this paper. Furthermore, we discuss the HQS partner structures of $T_{cc}$ and $T_{bb}$ obtained in our study by introducing the light cloud basis.
|
[
{
"version": "v1",
"created": "Fri, 14 Mar 2025 06:50:55 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 05:36:05 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Sakai",
"Manato",
""
],
[
"Yamaguchi",
"Yasuhiro",
""
]
] |
2503.11344
|
Dylan Pare Dr
|
Dylan M. Par\'e, David T. Chuss, Kaitlyn Karpovich, Natalie Butterfield, Edward J. Wollack, Mark R. Morris, Jeffrey Inara Iuliano
|
The Complex Magnetic Field of the Extreme Galactic Center: PRIMA Science Potential
|
Published in JATIS. This paper is part of the JATIS special issue focused on the PRobe Infrared Mission for Astrophysics (PRIMA) probe mission concept. The issue is edited by Matt Griffin and Naseem Rangwala (JATIS VOL. 11 & NO. 3 | July 2025)
| null |
10.1117/1.JATIS.11.3.031604
| null |
astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
The Central Molecular Zone (CMZ) of the Galactic Center (GC) region of the Milky Way contains a substantial fraction of the molecular mass of the Galaxy >10e7 solar masses yet exhibits an order of magnitude lower star formation efficiency (SFE) than expected given the high densities found in this region. There are multiple possible explanations for the depressed SFE in the CMZ, like feedback, strong turbulence, longer free-fall timescales, and high magnetic field strengths. It is currently unclear which of these mechanisms is the dominant inhibitor of star formation in the CMZ. It is important to understand the star formation process in the extreme environment of the CMZ because it is the only Galactic nuclear region we are able to study at high spatial resolutions with current observatories. One way to determine the relative importance of the different SFE inhibiting mechanisms is through multi-spatial and multi-frequency polarimetric observations of the CMZ. Such observations will provide insight into the behavior of the magnetic field in this unique environment. These observations will complement radio observations of non-thermal structures revealing the magnetic field morphology and polarization. The PRobe far--Infrared Mission for Astrophysics (PRIMA) will be uniquely capable of contributing to such explorations by providing unique resolutions and frequencies for polarimetric observations. The PRIMAger instrument will yield polarimetric observations covering the wavelength range 80 -- 261 um with beam sizes ranging from 11 -- 28'', capabilities that complement existing and upcoming observatories.
|
[
{
"version": "v1",
"created": "Fri, 14 Mar 2025 12:19:00 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 23:33:42 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Paré",
"Dylan M.",
""
],
[
"Chuss",
"David T.",
""
],
[
"Karpovich",
"Kaitlyn",
""
],
[
"Butterfield",
"Natalie",
""
],
[
"Wollack",
"Edward J.",
""
],
[
"Morris",
"Mark R.",
""
],
[
"Iuliano",
"Jeffrey Inara",
""
]
] |
2503.11396
|
Hossein Vahid
|
Hossein Vahid, Jens-Uwe Sommer, and Abhinav Sharma
|
Collective Dynamics in Active Polar Polymer Assemblies
| null | null | null | null |
cond-mat.soft physics.bio-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Tangentially driven active polymers (TDAPs), model systems for motor-driven filaments, have been extensively studied in uniform activity fields. Here, we show that an activity gradient breaks fore-aft symmetry, generating net body forces that steer dimers, asters, and larger assemblies toward high-activity regions. Including temporal stochasticity softens the chains, allowing them to bend and wind around other filaments. Once several contacts are established, steric interlocking arrests relative motion and stabilizes the assembly into a hierarchically entangled cluster. These clusters persist for times far exceeding single-chain relaxation and do not appear under deterministic, temporally constant activity. Remarkably, such activity-induced gelation occurs even at polymer concentrations substantially lower than those typically required for passive chains. Our results reveal a new mechanism for activity-induced aggregation, providing new strategies for designing autonomous and reconfigurable microfluidic systems.
|
[
{
"version": "v1",
"created": "Fri, 14 Mar 2025 13:40:23 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Apr 2025 09:24:17 GMT"
},
{
"version": "v3",
"created": "Tue, 02 Sep 2025 12:01:23 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Vahid",
"Hossein",
""
],
[
"Sommer",
"Jens-Uwe",
""
],
[
"Sharma",
"Abhinav",
""
]
] |
2503.11401
|
Yasutaka Koga
|
Yasutaka Koga, Nobuyuki Asaka, Masashi Kimura, Kazumasa Okabayashi
|
Shadow formation in gravitational collapse: Redshift and blueshift by spacetime dynamics
|
44 pages, 18 figures
|
Phys. Rev. D 112, 044069 (2025)
|
10.1103/cbzn-pzrw
|
YITP-25-22, RUP-25-2
|
gr-qc
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
A black hole illuminated by a background light source is observed as a black hole shadow. For a black hole formed by gravitational collapse of a transmissive object, redshift of light due to the spacetime dynamics is expected to play a crucial role in the shadow formation. In this paper, we investigate the redshift of light caused by the spacetime dynamics. First, we consider a spherical shell model. We see that the collapse of the shell typically leads to the redshift of light, while blueshift can be also observed in some cases. This result suggests that a shadow image is generally formed in the late stage of the gravitational collapse of a transmissive object. Second, we propose a covariant formula for the redshift of light in a general, dynamical, and spherically symmetric spacetime. This formula relates the redshift to the energy-momentum tensor of the background spacetime and provides its intuitive interpretation with a Newtonian analogy. The redshift effect analyzed in this work is regarded as the integrated Sachs-Wolfe effect or the Rees-Sciama effect in gravitational collapse.
|
[
{
"version": "v1",
"created": "Fri, 14 Mar 2025 13:43:27 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 07:38:08 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Koga",
"Yasutaka",
""
],
[
"Asaka",
"Nobuyuki",
""
],
[
"Kimura",
"Masashi",
""
],
[
"Okabayashi",
"Kazumasa",
""
]
] |
2503.11448
|
Michael Kachelrie{\ss}
|
M. Kachelriess and E. Lammert
|
Cygnus X-3 as semi-hidden PeVatron
|
7 pages, 8 pdf figures; v2: minor clarifications, accepted in A&A
|
A&A 701, A22 (2025)
|
10.1051/0004-6361/202554701
| null |
astro-ph.HE
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
The high-mass X-ray binary Cygnus X-3 has been suggested for a long time to be a source of high-energy photons and neutrinos. In view of the increased sensitivity of current experiments, we examine the acceleration and interactions of high-energy cosmic rays (CRs) in this binary system, assuming that the compact object is a black hole. Using a test-particle approach in a Monte-Carlo framework, we employ as the basic CR acceleration mechanisms magnetic reconnection or second-order Fermi and diffusive shock acceleration. We find that in all three scenarios CRs can be accelerated beyond PeV energies. High-energy photons and neutrinos are produced as secondaries in photo-hadronic interactions of CRs on X-ray photons and in the scattering on gas from the wind of the companion star. Normalising the predicted photon flux to the excess flux observed by LHAASO at energies above PeV in the direction of Cygnus X-3, a CR acceleration efficiency of $10^{-3}$ is sufficient to power the required CR luminosity. Our results suggest that the PeV photon flux from Cygnus X-3 could be in a bright phase significantly increased relative to the average flux of the last years.
|
[
{
"version": "v1",
"created": "Fri, 14 Mar 2025 14:32:32 GMT"
},
{
"version": "v2",
"created": "Fri, 20 Jun 2025 14:38:46 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Kachelriess",
"M.",
""
],
[
"Lammert",
"E.",
""
]
] |
2503.11815
|
Kendall Thomas
|
Kendall L. Thomas and Jan Hannig
|
Movement Dynamics in Elite Female Soccer Athletes: The Quantile Cube Approach
|
40 pages, 6 tables, 13 figures; Poster presented and awarded 1st place at the 2024 Carnegie Mellon Sports Analytics Conference. Paper submitted to the Journal of Quantitative Analysis in Sports in February 2025; Revision submitted August 2025;
| null | null | null |
stat.ME
|
http://creativecommons.org/licenses/by/4.0/
|
This paper presents the quantile cube, a novel three-dimensional summary representation designed to analyze external load using GPS-derived movement data. While broadly applicable, we demonstrate its utility through an application to data from elite female soccer athletes across 23 matches. The quantile cube segments athlete movements into discrete quantiles of velocity, acceleration, and movement angle across match halves, providing a structured and interpretable framework to capture complex movement dynamics. Statistical analysis revealed significant differences in movement distributions between the first and second halves for individual athletes across all matches. Principal Component Analysis identified matches with unique movement dynamics, particularly at the start and end of the season. Dirichlet-multinomial regression further explored how factors such as athlete position, playing time, and match characteristics influenced movement profiles. Our analysis reveals external load variations over time and provides insights into performance optimization. The integration of these statistical techniques demonstrates the potential of data-driven strategies to enhance athlete monitoring and workload management in women's soccer.
|
[
{
"version": "v1",
"created": "Fri, 14 Mar 2025 19:01:39 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 20:05:31 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Thomas",
"Kendall L.",
""
],
[
"Hannig",
"Jan",
""
]
] |
2503.11830
|
Dominique Sugny
|
O. Fresse-Colson, S. Gu\'erin, Xi Chen and D. Sugny
|
Application of the Pontryagin Maximum Principle to the robust time-optimal control of two-level quantum systems
|
30 pages, 7 figures
|
Phys. Rev. A 112, 022618 (2025)
|
10.1103/rn5j-kjq3
| null |
quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
We study the time-optimal robust control of a two-level quantum system subjected to field inhomogeneities. We apply the Pontryagin Maximum Principle and we introduce a reduced space onto which the optimal dynamics is projected down. This reduction leads to a complete analytical derivation of the optimal solution in terms of elliptic functions and elliptic integrals. Necessary optimality conditions are then obtained for the original system. These conditions are verified numerically and lead to the optimal control protocol. Various examples, ranging from state-to-state transfer to the generation of a Not gate, illustrate this study. The connection with other geometric optimization approaches that have been used to solve this problem is also discussed.
|
[
{
"version": "v1",
"created": "Fri, 14 Mar 2025 19:47:08 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Fresse-Colson",
"O.",
""
],
[
"Guérin",
"S.",
""
],
[
"Chen",
"Xi",
""
],
[
"Sugny",
"D.",
""
]
] |
2503.11999
|
Tongxuan Tian
|
Tongxuan Tian, Haoyang Li, Bo Ai, Xiaodi Yuan, Zhiao Huang, Hao Su
|
Diffusion Dynamics Models with Generative State Estimation for Cloth Manipulation
|
CoRL 2025. Project website: https://uniclothdiff.github.io/
| null | null | null |
cs.RO cs.CV cs.SY eess.SY
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Cloth manipulation is challenging due to its highly complex dynamics, near-infinite degrees of freedom, and frequent self-occlusions, which complicate both state estimation and dynamics modeling. Inspired by recent advances in generative models, we hypothesize that these expressive models can effectively capture intricate cloth configurations and deformation patterns from data. Therefore, we propose a diffusion-based generative approach for both perception and dynamics modeling. Specifically, we formulate state estimation as reconstructing full cloth states from partial observations and dynamics modeling as predicting future states given the current state and robot actions. Leveraging a transformer-based diffusion model, our method achieves accurate state reconstruction and reduces long-horizon dynamics prediction errors by an order of magnitude compared to prior approaches. We integrate our dynamics models with model predictive control and show that our framework enables effective cloth folding on real robotic systems, demonstrating the potential of generative models for deformable object manipulation under partial observability and complex dynamics.
|
[
{
"version": "v1",
"created": "Sat, 15 Mar 2025 05:34:26 GMT"
},
{
"version": "v2",
"created": "Fri, 29 Aug 2025 22:59:16 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Tian",
"Tongxuan",
""
],
[
"Li",
"Haoyang",
""
],
[
"Ai",
"Bo",
""
],
[
"Yuan",
"Xiaodi",
""
],
[
"Huang",
"Zhiao",
""
],
[
"Su",
"Hao",
""
]
] |
2503.12225
|
Rinku Dewri
|
Rinku Dewri
|
Interpretation Gaps in LLM-Assisted Comprehension of Privacy Documents
|
Minor revision
|
IEEE Computer, 58:9, pp. 70-79, 2025
|
10.1109/MC.2025.3575345
| null |
cs.CL cs.CY cs.IR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This article explores the gaps that can manifest when using a large language model (LLM) to obtain simplified interpretations of data practices from a complex privacy policy. We exemplify these gaps to showcase issues in accuracy, completeness, clarity and representation, while advocating for continued research to realize an LLM's true potential in revolutionizing privacy management through personal assistants and automated compliance checking.
|
[
{
"version": "v1",
"created": "Sat, 15 Mar 2025 18:43:13 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 00:59:58 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Dewri",
"Rinku",
""
]
] |
2503.12345
|
Zhongyuan Wang
|
Zhongyuan Wang, Richong Zhang, Zhijie Nie, Hangyu Mao
|
General Table Question Answering via Answer-Formula Joint Generation
|
work in progress
| null | null | null |
cs.CL cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Advanced table question answering (TableQA) methods prompt large language models (LLMs) to generate answer text, SQL query, Python code, or custom operation, which impressively improve the complex reasoning problems in the TableQA task. However, these methods lack the versatility to cope with specific question types or table structures. In contrast, the Spreadsheet Formula, the widely used and well-defined operation language for tabular data, has not been thoroughly explored to solve TableQA. In this paper, we first attempt to use the Formula as the executable representation for solving complex reasoning on tables with different structures. Specifically, we construct \texttt{FromulaQA}, a large Formula-annotated TableQA dataset from existing datasets. In addition, we propose \texttt{TabAF}, a general table answering framework to solve multiple types of tasks over multiple types of tables simultaneously, which decodes answers and Formulas with a single LLM backbone. Extensive experiments demonstrate the versatility and generalization of \texttt{TabAF}. Under the same model size, \texttt{TabAF} achieves new state-of-the-art performance on the WikiTableQuestion, HiTab, and TabFact.
|
[
{
"version": "v1",
"created": "Sun, 16 Mar 2025 03:51:06 GMT"
},
{
"version": "v2",
"created": "Fri, 23 May 2025 18:02:35 GMT"
},
{
"version": "v3",
"created": "Sun, 31 Aug 2025 14:11:57 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Wang",
"Zhongyuan",
""
],
[
"Zhang",
"Richong",
""
],
[
"Nie",
"Zhijie",
""
],
[
"Mao",
"Hangyu",
""
]
] |
2503.12569
|
Enrico Schiappacasse
|
Enrico D. Schiappacasse
|
Dark Spin-2 Field Solitons as a Source of Electromagnetic Radiation
|
44 pages, 7 figures; v2 - references added, updated towards version published in Journal of Cosmology and Astroparticle Physics (JCAP)
|
JCAP 08 (2025) 085
|
10.1088/1475-7516/2025/08/085
| null |
hep-ph astro-ph.CO astro-ph.HE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Spin-$s$ light dark boson particles can exhibit wave-like behavior, capable of forming long-lived, coherent, spatially localized structures known as solitons. This work considers the possibility that a light spin-2 particle might be part of or all the dark matter content of the Universe, which could result in a significant fraction of solitons existing today in galactic halos. If these dark matter particles interact with electromagnetism through dimension-6 operators, the solitons may experience parametric resonance of photons triggered by the surrounding electromagnetic field. We explore the feasibility and key characteristics of this electromagnetic radiation, as well as the potential for detection through soliton mergers using ground-based facilities.
|
[
{
"version": "v1",
"created": "Sun, 16 Mar 2025 16:56:53 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 16:43:35 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Schiappacasse",
"Enrico D.",
""
]
] |
2503.12608
|
Andrei-Marius Avram
|
Andrei-Marius Avram, Marian Lupa\c{s}cu, Dumitru-Clementin Cercel, Ionu\c{t} Mironic\u{a} and \c{S}tefan Tr\u{a}u\c{s}an-Matu
|
UniBERT: Adversarial Training for Language-Universal Representations
| null | null | null | null |
cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
This paper presents UniBERT, a compact multilingual language model that uses an innovative training framework that integrates three components: masked language modeling, adversarial training, and knowledge distillation. Pre-trained on a meticulously curated Wikipedia corpus spanning 107 languages, UniBERT is designed to reduce the computational demands of large-scale models while maintaining competitive performance across various natural language processing tasks. Comprehensive evaluations on four tasks - named entity recognition, natural language inference, question answering, and semantic textual similarity - demonstrate that our multilingual training strategy enhanced by an adversarial objective significantly improves cross-lingual generalization. Specifically, UniBERT models show an average relative improvement of 7.72% over traditional baselines, which achieved an average relative improvement of only 1.17%, and statistical analysis confirms the significance of these gains (p-value = 0.0181). This work highlights the benefits of combining adversarial training and knowledge distillation to build scalable and robust language models, thus advancing the field of multilingual and cross-lingual natural language processing.
|
[
{
"version": "v1",
"created": "Sun, 16 Mar 2025 18:44:06 GMT"
},
{
"version": "v2",
"created": "Tue, 08 Apr 2025 18:40:25 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 18:01:37 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Avram",
"Andrei-Marius",
""
],
[
"Lupaşcu",
"Marian",
""
],
[
"Cercel",
"Dumitru-Clementin",
""
],
[
"Mironică",
"Ionuţ",
""
],
[
"Trăuşan-Matu",
"Ştefan",
""
]
] |
2503.12642
|
Anjali Dharmik
|
Anjali Dharmik
|
COVID-19 Pneumonia Diagnosis Using Medical Images: Deep Learning-Based Transfer Learning Approach
| null | null | null | null |
eess.IV cs.AI cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
SARS-CoV-2, the causative agent of COVID-19, remains a global health concern due to its high transmissibility and evolving variants. Although vaccination efforts and therapeutic advancements have mitigated disease severity, emerging mutations continue to challenge diagnostics and containment strategies. As of mid-February 2025, global test positivity has risen to 11%, marking the highest level in over six months despite widespread immunization efforts. Newer variants demonstrate enhanced host cell binding, increasing both infectivity and diagnostic complexity. This study evaluates the effectiveness of deep transfer learning in delivering rapid, accurate, and mutation-resilient COVID-19 diagnosis from medical imaging, with a focus on scalability and accessibility. We developed an automated detection system using state-of-the-art CNNs, including VGG16, ResNet50, ConvNetXtTiny, MobileNet, NASNetMobile, and DenseNet121 among others, to detect COVID-19 from chest X-ray and CT images. Among all the models evaluated, DenseNet121 emerged as the best-performing architecture for COVID-19 diagnosis using CT and X-ray images. It achieved an impressive accuracy of 98%, with 96.9% precision, 98.9% recall, 97.9% F1-score and 99.8% AUC score, indicating a high degree of consistency and reliability in both detecting positive and negative cases. The confusion matrix showed minimal false positives and false negatives, underscoring the model's robustness in real-world diagnostic scenarios.
|
[
{
"version": "v1",
"created": "Sun, 16 Mar 2025 20:33:39 GMT"
},
{
"version": "v2",
"created": "Sun, 23 Mar 2025 17:38:40 GMT"
},
{
"version": "v3",
"created": "Sun, 13 Jul 2025 23:28:48 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Dharmik",
"Anjali",
""
]
] |
2503.12643
|
Ryan Caverly
|
Ping-Yen Shen and Ryan J. Caverly
|
Mass Translation and Reflectivity Device Solar Sail Momentum Management Using Predictive Control
|
Submitted to the Journal of Guidance, Control, and Dynamics
| null | null | null |
physics.space-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Solar sails enable propellant-free space missions by utilizing solar radiation pressure as thrust. However, disturbance torques act on the solar sail and effective attitude control leads to the continuous accumulation of reaction wheel angular momentum, necessitating an efficient momentum management strategy to prevent saturation. This paper presents a novel momentum management controller using model predictive control (MPC) that is tailored for solar sails, accommodating the unique actuation mechanisms of an active mass translator (AMT) and reflectivity control devices (RCDs). A first-order hold discretization and tailored motion costs are applied to the AMT translation, while the RCD actuation is handled using pulse-width modulation (PWM)-inspired quantization to address their on-off inputs. To enhance prediction accuracy, an iterative backwards-in-time MPC approach is introduced, incorporating the effects of PWM-quantized inputs into the optimization process. The dynamic model accounts for the time-dependent center of mass and moment of inertia changes caused by AMT translation, extending its applicability to other spacecraft with mass-shifting actuators. Simulation results demonstrate the effectiveness of the proposed framework in reaction wheel desaturation, attitude control, and momentum management actuation efficiency, highlighting the potential of integrating MPC to manage coupled nonlinear dynamics and discrete actuator constraints for solar sails.
|
[
{
"version": "v1",
"created": "Sun, 16 Mar 2025 20:38:23 GMT"
},
{
"version": "v2",
"created": "Fri, 29 Aug 2025 19:52:46 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Shen",
"Ping-Yen",
""
],
[
"Caverly",
"Ryan J.",
""
]
] |
2503.12655
|
Stefano Vergani
|
Callum Duffy, Marcin Jastrzebski, Stefano Vergani, Leigh H. Whitehead, Ryan Cross, Andrew Blake, Sarah Malik, John Marshall
|
LArTPC hit-based topology classification with quantum machine learning and symmetry
| null | null | null | null |
physics.ins-det hep-ex physics.data-an
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present a new approach to separate track-like and shower-like topologies in liquid argon time projection chamber (LArTPC) experiments for neutrino physics using quantum machine learning. Effective reconstruction of neutrino events in LArTPCs requires accurate and granular information about the energy deposited in the detector. These energy deposits can be viewed as 2-D images. Simulated data from the MicroBooNE experiment and a simple custom dataset are used to perform pixel-level classification of the underlying particle topology. Images of the events have been studied by creating small patches around each pixel to characterise its topology based on its immediate neighbourhood. This classification is achieved using convolution-based learning models, including quantum-enhanced architectures known as quanvolutional neural networks. The quanvolutional networks are extended to symmetries beyond translation. Rotational symmetry has been incorporated into a subset of the models. Quantum-enhanced models perform better than their classical counterparts with a comparable number of parameters but are outperformed by classical models, which contain an order of magnitude more parameters. The inclusion of rotation symmetry appears to benefit only large models and remains to be explored further.
|
[
{
"version": "v1",
"created": "Sun, 16 Mar 2025 21:20:42 GMT"
},
{
"version": "v2",
"created": "Thu, 15 May 2025 15:13:44 GMT"
},
{
"version": "v3",
"created": "Mon, 19 May 2025 16:55:13 GMT"
},
{
"version": "v4",
"created": "Thu, 24 Jul 2025 13:25:17 GMT"
},
{
"version": "v5",
"created": "Sun, 31 Aug 2025 22:06:31 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Duffy",
"Callum",
""
],
[
"Jastrzebski",
"Marcin",
""
],
[
"Vergani",
"Stefano",
""
],
[
"Whitehead",
"Leigh H.",
""
],
[
"Cross",
"Ryan",
""
],
[
"Blake",
"Andrew",
""
],
[
"Malik",
"Sarah",
""
],
[
"Marshall",
"John",
""
]
] |
2503.12962
|
Abolfazl Zakeri
|
Abolfazl Zakeri, Mohammad Moltafet, and Marian Codreanu
|
Goal-Oriented Remote Tracking Through Correlated Observations in Pull-based Communications
|
This is a full version of an IEEE COML paper (under revision)
| null | null | null |
eess.SP
|
http://creativecommons.org/licenses/by-sa/4.0/
|
We address the real-time remote tracking problem in a status update system comprising two sensors, two independent information sources, and a remote monitor. The status updating follows a pull-based communication, where the monitor commands/pulls the sensors for status updates, i.e., the actual state of the sources. We consider that the observations are \textit{correlated}, meaning that each sensor's sent data could also include the state of the other source due to, e.g., inter-sensor communications or overlapping monitoring regions. The effectiveness of data communication is measured by a generic distortion, capturing the underlying application goal. We provide optimal command/pulling policies for the monitor that minimize the average weighted sum distortion and transmission cost. Since the monitor cannot fully observe the exact state of each source, we propose a partially observable Markov decision process (POMDP) and reformulate it as a belief MDP problem. We then effectively truncate the infinite belief space and transform it into a finite-state MDP problem, which is solved via relative value iteration. Simulation results show the effectiveness of the derived policy over age-based and deep-Q network baseline policies.
|
[
{
"version": "v1",
"created": "Mon, 17 Mar 2025 09:17:31 GMT"
},
{
"version": "v2",
"created": "Mon, 14 Jul 2025 17:14:34 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 07:14:46 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Zakeri",
"Abolfazl",
""
],
[
"Moltafet",
"Mohammad",
""
],
[
"Codreanu",
"Marian",
""
]
] |
2503.13048
|
Haofeng Chen
|
Haofeng Chen, Bedrich Himmel, Bin Li, Xiaojie Wang, Matej Hoffmann
|
Multi-Touch and Bending Perception Using Electrical Impedance Tomography for Robotics
|
This work has been submitted to the IEEE for possible publication
| null | null | null |
cs.RO
|
http://creativecommons.org/licenses/by/4.0/
|
Electrical Impedance Tomography (EIT) offers a promising solution for distributed tactile sensing with minimal wiring and full-surface coverage in robotic applications. However, EIT-based tactile sensors face significant challenges during surface bending. Deformation alters the baseline impedance distribution and couples with touch-induced conductivity variations, complicating signal interpretation. To address this challenge, we present a novel sensing framework that integrates a deep neural network for interaction state classification with a dynamic adaptive reference strategy to decouple touch and deformation signals, while a data-driven regression model translates EIT voltage changes into continuous bending angles. The framework is validated using a magnetic hydrogel composite sensor that conforms to bendable surfaces. Experimental evaluations demonstrate that the proposed framework achieves precise and robust bending angle estimation, high accuracy in distinguishing touch, bending, and idle states, and significantly improves touch localization quality under bending deformation compared to conventional fixed-reference methods. Real-time experiments confirm the system's capability to reliably detect multi-touch interactions and track bending angles across varying deformation conditions. This work paves the way for flexible EIT-based robotic skins capable of rich multimodal sensing in robotics and human-robot interaction.
|
[
{
"version": "v1",
"created": "Mon, 17 Mar 2025 10:53:42 GMT"
},
{
"version": "v2",
"created": "Tue, 26 Aug 2025 06:43:58 GMT"
},
{
"version": "v3",
"created": "Tue, 02 Sep 2025 15:47:35 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Chen",
"Haofeng",
""
],
[
"Himmel",
"Bedrich",
""
],
[
"Li",
"Bin",
""
],
[
"Wang",
"Xiaojie",
""
],
[
"Hoffmann",
"Matej",
""
]
] |
2503.13150
|
Samuel Cameron
|
Samuel Cameron and Elsen Tjhung
|
Exact Results in Stochastic Processes with Division, Death, and Diffusion: Spatial Correlations, Marginal Entropy Production, and Macroscopic Currents
|
18 pages, 8 figures
| null |
10.1103/b9n6-mynw
| null |
cond-mat.stat-mech cond-mat.soft physics.bio-ph
|
http://creativecommons.org/licenses/by/4.0/
|
We consider a generic class of stochastic particle-based models whose state at an instant in time is described by a set of continuous degrees of freedom (e.g. positions), and the length of this set changes stochastically in time due to birth-death processes. Using a master equation formalism, we write down the dynamics of the corresponding (infinite) set of probability distributions: this takes the form of coupled Fokker-Planck equations with model-dependent source and sink terms. We derive the general expression of entropy production rate for this class of models in terms of path irreversibility. To demonstrate the practical use of this framework, we analyze a biologically motivated model incorporating division, death, and diffusion, where spatial correlations arise through the division process. By systematically integrating out excess degrees of freedom, we obtain the marginal probability distribution, enabling exact calculations of key statistical properties such as average density and correlation functions. We validate our analytical results through numerical Brownian dynamics simulations, finding excellent agreement between theory and simulation. Our method thus provides a powerful tool for tackling previously unsolved problems in stochastic birth-death dynamics.
|
[
{
"version": "v1",
"created": "Mon, 17 Mar 2025 13:20:53 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 14:35:51 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Cameron",
"Samuel",
""
],
[
"Tjhung",
"Elsen",
""
]
] |
2503.13158
|
Bernd Zimmering
|
Bernd Zimmering, Cec\'ilia Coelho, Vaibhav Gupta, Maria Maleshkova, Oliver Niggemann
|
Breaking Free: Decoupling Forced Systems with Laplace Neural Networks
|
Preprint - Accepted to the Research Track of ECML PKDD 2025
| null | null | null |
cs.LG cs.SY eess.SY
|
http://creativecommons.org/licenses/by/4.0/
|
Modelling forced dynamical systems - where an external input drives the system state - is critical across diverse domains such as engineering, finance, and the natural sciences. In this work, we propose Laplace-Net, a decoupled, solver-free neural framework for learning forced and delay-aware systems. It leverages a Laplace transform-based approach to decompose internal dynamics, external inputs, and initial values into established theoretical concepts, enhancing interpretability. Laplace-Net promotes transferability since the system can be rapidly re-trained or fine-tuned for new forcing signals, providing flexibility in applications ranging from controller adaptation to long-horizon forecasting. Experimental results on eight benchmark datasets - including linear, non-linear, and delayed systems - demonstrate the method's improved accuracy and robustness compared to state-of-the-art approaches, particularly in handling complex and previously unseen inputs.
|
[
{
"version": "v1",
"created": "Mon, 17 Mar 2025 13:31:12 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 10:37:52 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Zimmering",
"Bernd",
""
],
[
"Coelho",
"Cecília",
""
],
[
"Gupta",
"Vaibhav",
""
],
[
"Maleshkova",
"Maria",
""
],
[
"Niggemann",
"Oliver",
""
]
] |
2503.13247
|
Shaokai Jian
|
Huan Jiang, Yang Ge, Shao-Kai Jian
|
Boundary criticality for the Gross-Neveu-Yukawa models
|
4.5 pages + Supplemental Material, 2 figures, added new determinant quantum Monte Carlo simulations
| null | null | null |
cond-mat.str-el cond-mat.stat-mech hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
We study the boundary criticality for the Gross-Neveu-Yukawa (GNY) models. Employing interacting Dirac fermions on a honeycomb lattice with armchair boundaries, we use determinate quantum Monte Carlo (DQMC) simulation to uncover rich boundary criticalities at the quantum phase transition to a charge-density-wave (CDW) insulator, including the ordinary, special, and extraordinary transitions. The Dirac fermions satisfy a Dirichlet boundary condition, while the boson field, representing the CDW order, obeys Dirichlet and Neumann conditions at the ordinary and special transitions, respectively, thereby enriching the critical GNY model. We develop a perturbative $4-\epsilon$ renormalization group approach to compute the boundary critical exponents. Our framework generalizes to other GNY universality class variants and provides theoretical predictions for experiments.
|
[
{
"version": "v1",
"created": "Mon, 17 Mar 2025 15:01:25 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 21:17:58 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Jiang",
"Huan",
""
],
[
"Ge",
"Yang",
""
],
[
"Jian",
"Shao-Kai",
""
]
] |
2503.13669
|
Jonas Santos F. G.
|
Jonas F. G. Santos
|
Enhanced frequency and temperature estimation by a $\mathcal{PT}$-symmetric quantum oscillator
| null |
Phys. Rev. A 112, 012220 (2025)
|
10.1103/xdnc-tc2y
| null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Quantum metrology employs quantum properties to enhance the precision of physical parameters, in order to characterize quantum states as well as channels. Frequency and temperature estimations are of fundamental importance for these tasks and have been considerably treated in quantum sensing strategies. From the set of quantum features that can be exploited in quantum metrology, those related to non-Hermitian systems have received much attention in recent times. Here, we consider a paradigmatic non-Hermitian system, the quantum Swanson oscillator, as a probe system, in order to investigate how the non-Hermitian contribution affects the frequency and temperature estimation. We use the quantum Fisher information to compute the main results. Furthermore, to perform a fair comparison, we define a gain function which is the ratio between the quantum Fisher information with the non-Hermitian contribution and the quantum Fisher information when this term becomes Hermitian. In addition, the quantum Fisher information is discussed in terms of the energetic cost to include the non-Hermitian contribution. Given that we obtain the Hermitian counterpart of the Hamiltonian by applying the Dyson map, we also study the estimation of the parameters characterizing this mapping. Our results indicate that the non-Hermitian contribution in the Swanson quantum oscillator can contribute to enhance the frequency and temperature estimation.
|
[
{
"version": "v1",
"created": "Mon, 17 Mar 2025 19:17:04 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Santos",
"Jonas F. G.",
""
]
] |
2503.13753
|
Avi Kadria
|
Avi Kadria, Liam Roditty
|
Compact routing schemes in undirected and directed graphs
| null | null | null | null |
cs.NI
|
http://creativecommons.org/licenses/by/4.0/
|
In this paper, we study the problem of compact routing schemes in weighted undirected and directed graphs.
\textit{For weighted undirected graphs}, more than a decade ago, Chechik [PODC'13] presented a $\approx3.68k$-stretch compact routing scheme that uses $\tilde{O}(n^{1/k}\log{D})$ local storage, where $D$ is the normalized diameter, for every $k>1$. We present a $\approx 2.64k$-stretch compact routing scheme that uses $\tilde{O}(n^{1/k})$ local storage \textit{on average} in each vertex. This is the first compact routing scheme that uses total local storage of $\tilde{O}(n^{1+1/k})$ while achieving a $c \cdot k$ stretch, for a constant $c < 3$.
In real-world network protocols, messages are usually transformed as part of a communication session between two parties. Therefore, more than two decades ago, Thorup and Zwick [SPAA'01] considered compact routing schemes that establish a communication session using a handshake. In their handshake-based compact routing scheme, the handshake is routed along a $(4k-5)$-stretch path, and the rest of the communication session is routed along an optimal $(2k-1)$-stretch path. It is straightforward to improve the $(4k-5)$-stretch of the handshake to $\approx3.68k$-stretch using the compact routing scheme of Chechik [PODC'13]. We improve the handshake stretch to the optimal $(2k-1)$, by borrowing the concept of roundtrip routing from directed graphs to \textit{undirected} graphs.
\textit{For weighted directed graphs}, more than two decades ago, Roditty, Thorup, and Zwick [SODA'02 and TALG'08] presented a $(4k+\eps)$-stretch compact roundtrip routing scheme that uses $\tilde{O}(n^{1/k})$ local storage for every $k\ge 3$. For $k=3$, this gives a $(12+\eps)$-roundtrip stretch using $\tilde{O}(n^{1/3})$ local storage. We improve the stretch by developing a $7$-roundtrip stretch routing scheme with $\tilde{O}(n^{1/3})$ local storage.
|
[
{
"version": "v1",
"created": "Mon, 17 Mar 2025 22:20:32 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 08:55:50 GMT"
},
{
"version": "v3",
"created": "Sun, 31 Aug 2025 09:23:32 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Kadria",
"Avi",
""
],
[
"Roditty",
"Liam",
""
]
] |
2503.13932
|
Xinze Zhang
|
Xinze Zhang and Yong Li
|
Most Probable KAM Tori in Stochastic Hamiltonian Systems
|
39 pages, 4 figures
| null | null | null |
math.DS math.CA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper conducts an in-depth investigation into the impact of stochastic perturbations-particularly multiplicative noise-on the integrable structures of Hamiltonian systems, with a central focus on developing a KAM theory for stochastic Hamiltonian dynamics. We begin by deriving the Onsager-Machlup functional for Hamiltonian systems driven by multiplicative noise and identifying the most probable transition paths of system trajectories. This analysis reveals the fundamental differences in how additive versus multiplicative noise influences the integrability of Hamiltonian systems. Building upon this, we establish a large deviation principle for the system and derive a rate function that quantitatively characterizes trajectory deviations, especially in the regime of rare events. The main contribution of this work lies in demonstrating that, under the small noise limit, the quasi-periodic invariant tori of the unperturbed system persist in a probabilistic sense, indicating the stability of KAM structures under stochastic perturbations. Furthermore, we show that the exponential rate of deviation from the invariant tori exactly matches the large deviation rate function, thus providing a quantitative characterization of the structural persistence and fluctuation geometry of quasi-periodic motions in stochastic Hamiltonian systems. These results extend the classical KAM framework to a stochastic setting and offer new insights into the behavior of complex dynamical systems under the influence of noise.
|
[
{
"version": "v1",
"created": "Tue, 18 Mar 2025 05:59:32 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 02:47:45 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Zhang",
"Xinze",
""
],
[
"Li",
"Yong",
""
]
] |
2503.14093
|
Melanie Habermann
|
Melanie Habermann, Ashkaan K. Fahimipour, Justin D. Yeakel, and Thilo Gross
|
Functional Motifs in Foodwebs and Networks
|
Version 2: Corrected figures (Fig. 3, Fig. 4, and Fig. S1) and corrected percentages in the discussion of the results in the part "Functional Reactivity Motifs" after a bug was discovered in the code. Main part: 17 pages, 4 figures; supporting information: 13 pages, 1 figure
| null | null | null |
q-bio.PE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
When studying a complex system it is often useful to think of the system as a network of interacting units. One can then ask if some properties of the entire network are already explained by a small part of the network - a network motif. A famous example of an ecological motif is competitive exclusion in foodwebs, where the presence of two species competing for a shared resource precludes the existence of a stable equilibrium for the whole system. However, other examples of motifs with such direct impacts on stability are not known. Here we explain why small motifs that allow conclusions on systemic stability are rare. Moreover, we show that another dynamical property, reactivity, is naturally rooted in motifs. Computing the reactivity in motifs can reveal which parts of a network are prone to respond violently to perturbations. This highlights motif reactivity as a useful property to measure in real-world systems to understand likely modes of systemic failure in foodwebs, epidemics, supply chains and other applications.
|
[
{
"version": "v1",
"created": "Tue, 18 Mar 2025 10:11:09 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 20:47:32 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Habermann",
"Melanie",
""
],
[
"Fahimipour",
"Ashkaan K.",
""
],
[
"Yeakel",
"Justin D.",
""
],
[
"Gross",
"Thilo",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.