id
stringlengths 9
16
| submitter
stringlengths 1
64
⌀ | authors
stringlengths 3
62.8k
| title
stringlengths 1
382
| comments
stringlengths 1
1.15k
⌀ | journal-ref
stringlengths 1
557
⌀ | doi
stringlengths 7
153
⌀ | report-no
stringlengths 1
509
⌀ | categories
stringlengths 5
125
| license
stringclasses 9
values | abstract
stringlengths 3
6.09k
| versions
listlengths 1
187
| update_date
timestamp[s]date 2007-05-23 00:00:00
2025-08-29 00:00:00
| authors_parsed
listlengths 1
3.31k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2508.20737
|
Wei Ma
|
Wei Ma and Yixiao Yang and Qiang Hu and Shi Ying and Zhi Jin and Bo Du and Zhenchang Xing and Tianlin Li and Junjie Shi and Yang Liu and Linxiao Jiang
|
Rethinking Testing for LLM Applications: Characteristics, Challenges, and a Lightweight Interaction Protocol
| null | null | null | null |
cs.SE cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Applications of Large Language Models~(LLMs) have evolved from simple text generators into complex software systems that integrate retrieval augmentation, tool invocation, and multi-turn interactions. Their inherent non-determinism, dynamism, and context dependence pose fundamental challenges for quality assurance. This paper decomposes LLM applications into a three-layer architecture: \textbf{\textit{System Shell Layer}}, \textbf{\textit{Prompt Orchestration Layer}}, and \textbf{\textit{LLM Inference Core}}. We then assess the applicability of traditional software testing methods in each layer: directly applicable at the shell layer, requiring semantic reinterpretation at the orchestration layer, and necessitating paradigm shifts at the inference core. A comparative analysis of Testing AI methods from the software engineering community and safety analysis techniques from the AI community reveals structural disconnects in testing unit abstraction, evaluation metrics, and lifecycle management. We identify four fundamental differences that underlie 6 core challenges. To address these, we propose four types of collaborative strategies (\emph{Retain}, \emph{Translate}, \emph{Integrate}, and \emph{Runtime}) and explore a closed-loop, trustworthy quality assurance framework that combines pre-deployment validation with runtime monitoring. Based on these strategies, we offer practical guidance and a protocol proposal to support the standardization and tooling of LLM application testing. We propose a protocol \textbf{\textit{Agent Interaction Communication Language}} (AICL) that is used to communicate between AI agents. AICL has the test-oriented features and is easily integrated in the current agent framework.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:00:28 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Ma",
"Wei",
""
],
[
"Yang",
"Yixiao",
""
],
[
"Hu",
"Qiang",
""
],
[
"Ying",
"Shi",
""
],
[
"Jin",
"Zhi",
""
],
[
"Du",
"Bo",
""
],
[
"Xing",
"Zhenchang",
""
],
[
"Li",
"Tianlin",
""
],
[
"Shi",
"Junjie",
""
],
[
"Liu",
"Yang",
""
],
[
"Jiang",
"Linxiao",
""
]
] |
2508.20738
|
Lukas Bartl
|
Lukas Bartl, Jasmin Blanchette, Tobias Nipkow
|
Exploiting Instantiations from Paramodulation Proofs in Isabelle/HOL
|
This version of the contribution has been accepted for publication in CADE-30, after peer review but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections
| null |
10.1007/978-3-031-99984-0_30
| null |
cs.LO
|
http://creativecommons.org/licenses/by/4.0/
|
Metis is an ordered paramodulation prover built into the Isabelle/HOL proof assistant. It attempts to close the current goal using a given list of lemmas. Typically these lemmas are found by Sledgehammer, a tool that integrates external automatic provers. We present a new tool that analyzes successful Metis proofs to derive variable instantiations. These increase Sledgehammer's success rate, improve the speed of Sledgehammer-generated proofs, and help users understand why a goal follows from the lemmas.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:00:35 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Bartl",
"Lukas",
""
],
[
"Blanchette",
"Jasmin",
""
],
[
"Nipkow",
"Tobias",
""
]
] |
2508.20739
|
Davide Perego
|
Davide Perego and Matteo Tarocchi
|
Limit spaces of vertex and edge replacement systems
|
28 pages, comments are welcome
| null | null | null |
math.CO math.DS math.GR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We introduce and study VERSs (vertex and edge replacement systems) as a technology of graph expansions. We consider its history graph, an augmented tree that records each graph expansion, and we provide sufficient conditions under which it is hyperbolic. When hyperbolic, its Gromov boundary is what we call the limit space of the VERS. We provide three examples from different areas of mathematics: Schreier graphs and limit spaces of finitely generated contracting self-similar groups, injective post-critically finite iterated function systems and limit spaces of edge replacement systems.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:03:04 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Perego",
"Davide",
""
],
[
"Tarocchi",
"Matteo",
""
]
] |
2508.20740
|
Yuki Tanaka
|
Yuki Tanaka and Seiichiro Katsura
|
Non-expert to Expert Motion Translation Using Generative Adversarial Networks
| null | null | null | null |
cs.RO cs.SY eess.SY
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Decreasing skilled workers is a very serious problem in the world. To deal with this problem, the skill transfer from experts to robots has been researched. These methods which teach robots by human motion are called imitation learning. Experts' skills generally appear in not only position data, but also force data. Thus, position and force data need to be saved and reproduced. To realize this, a lot of research has been conducted in the framework of a motion-copying system. Recent research uses machine learning methods to generate motion commands. However, most of them could not change tasks by following human intention. Some of them can change tasks by conditional training, but the labels are limited. Thus, we propose the flexible motion translation method by using Generative Adversarial Networks. The proposed method enables users to teach robots tasks by inputting data, and skills by a trained model. We evaluated the proposed system with a 3-DOF calligraphy robot.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:03:30 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Tanaka",
"Yuki",
""
],
[
"Katsura",
"Seiichiro",
""
]
] |
2508.20741
|
Chenhao Zhang
|
Chenhao Zhang, Wei Gao
|
AdaDPCC: Adaptive Rate Control and Rate-Distortion-Complexity Optimization for Dynamic Point Cloud Compression
| null | null | null | null |
cs.MM
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Dynamic point cloud compression (DPCC) is crucial in applications like autonomous driving and AR/VR. Current compression methods face challenges with complexity management and rate control. This paper introduces a novel dynamic coding framework that supports variable bitrate and computational complexities. Our approach includes a slimmable framework with multiple coding routes, allowing for efficient Rate-Distortion-Complexity Optimization (RDCO) within a single model. To address data sparsity in inter-frame prediction, we propose the coarse-to-fine motion estimation and compensation module that deconstructs geometric information while expanding the perceptive field. Additionally, we propose a precise rate control module that content-adaptively navigates point cloud frames through various coding routes to meet target bitrates. The experimental results demonstrate that our approach reduces the average BD-Rate by 5.81% and improves the BD-PSNR by 0.42 dB compared to the state-of-the-art method, while keeping the average bitrate error at 0.40%. Moreover, the average coding time is reduced by up to 44.6% compared to D-DPCC, underscoring its efficiency in real-time and bitrate-constrained DPCC scenarios. Our code is available at https://git.openi.org.cn/OpenPointCloud/Ada_DPCC.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:03:33 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Zhang",
"Chenhao",
""
],
[
"Gao",
"Wei",
""
]
] |
2508.20742
|
Elzbieta Gradauskaite
|
Elzbieta Gradauskaite
|
Revival of Layered Ferroelectrics in Thin Films
|
Literature review, 37 pages, 14 figures
| null | null | null |
cond-mat.mtrl-sci cond-mat.mes-hall physics.chem-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Layered perovskites are a versatile class of ferroelectrics with their structural anisotropy reflected in unusual electrostatics that give rise to exceptional ferroelectric properties. These materials fall into four main families: Aurivillius, Carpy-Galy, Ruddlesden-Popper, and Dion-Jacobson phases; each forming natural superlattices by interleaving perovskite slabs with spacer layers. For a long time, these materials were considered too structurally complex to prepare as high-quality thin films. However, recent breakthroughs in deposition and advanced characterization have made it possible to stabilize high-quality films with atomic precision, uncovering a wide range of unconventional ferroelectric functionalities. These include robust in-plane polarization without a critical thickness, the emergence of charged domain walls and non-trivial polar textures, resilience to doping with magnetic ions and charge carriers, and possibility to epitaxially integrate them into standard perovskite heterostructures. This review aims to unify current knowledge on the fabrication and characterization of layered ferroelectric thin films, and to present research findings across all four structural families, with the goal of highlighting their common features despite differences in crystal structure and polarization mechanisms. We also discuss promising research directions, including polar metallicity, (alter-)magnetoelectricity, exfoliation, and soft-chemistry-driven phase transformations, with the goal of consolidating the field and encouraging further exploration of these materials for both fundamental studies and applications.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:03:39 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Gradauskaite",
"Elzbieta",
""
]
] |
2508.20743
|
Battulga Munkhbat
|
Pawe{\l} Wyborski and Athanasios Paralikis and Pietro Metuh and Martin A. Jacobsen and Christian Ruiz and Niels Gregersen and Battulga Munkhbat
|
Toward triggered generation of indistinguishable single-photons from MoTe$_2$ quantum emitters
| null | null | null | null |
physics.optics cond-mat.mes-hall cond-mat.mtrl-sci quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Single-photon sources operating in the telecom band are fundamental components for long-distance optical quantum communication and information processing. Two-dimensional (2D) transition metal dichalcogenides (TMDs) offer a promising platform for such sources, but their development has been hindered by limited spectral range and poor single-photon indistinguishability. Here, we demonstrate a reproducible and systematic approach for generating near-infrared (1090-1200 nm) quantum emitters in bilayer MoTe$_2$ using deterministic strain and defect engineering. These emitters exhibit strong linear polarization (DOLP $>70\%$), sub-nanosecond lifetimes ($\tau \leqslant$ ~450 ps), high single-photon purity ($g^{(2)}(0)<0.1$), and resolution-limited emission ($\sim$200 $\mu$eV). Electrostatic biasing enables Stark tuning over a $\sim$3 meV range, reduced photon bunching, and significantly shortened radiative lifetimes, yielding narrow emission with ratios of experimental to transform-limited linewidths as low as $R\sim55$. Most notably, two-photon interference measurements reveal a Hong-Ou-Mandel visibility of $V_\text{HOM}\sim $10$\%$, and up to $V_\text{HOM}\sim$ 40$\%$ with post-selection by temporal filtering, representing the highest reported indistinguishability for any TMD quantum emitters and the first such demonstration in the near-infrared regime. These results establish MoTe$_2$ as a viable platform for tunable, low-noise, high-purity single-photon sources with promising indistinguishability, paving the way for their integration into telecom-compatible quantum photonic technologies.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:04:01 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Wyborski",
"Paweł",
""
],
[
"Paralikis",
"Athanasios",
""
],
[
"Metuh",
"Pietro",
""
],
[
"Jacobsen",
"Martin A.",
""
],
[
"Ruiz",
"Christian",
""
],
[
"Gregersen",
"Niels",
""
],
[
"Munkhbat",
"Battulga",
""
]
] |
2508.20744
|
Mehrdad Sabetzadeh
|
Shabnam Hassani, Mehrdad Sabetzadeh, Daniel Amyot
|
From Law to Gherkin: A Human-Centred Quasi-Experiment on the Quality of LLM-Generated Behavioural Specifications from Food-Safety Regulations
| null | null | null | null |
cs.SE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Context: Laws and regulations increasingly affect software design and quality assurance, but legal texts are written in technology-neutral language. This creates challenges for engineers who must develop compliance artifacts such as requirements and acceptance criteria. Manual creation is labor-intensive, error-prone, and requires domain expertise. Advances in Generative AI (GenAI), especially Large Language Models (LLMs), offer a way to automate deriving such artifacts.
Objective: We present the first systematic human-subject study of LLMs' ability to derive behavioral specifications from legal texts using a quasi-experimental design. These specifications translate legal requirements into a developer-friendly form.
Methods: Ten participants evaluated specifications generated from food-safety regulations by Claude and Llama. Using Gherkin, a structured BDD language, 60 specifications were produced. Each participant assessed 12 across five criteria: Relevance, Clarity, Completeness, Singularity, and Time Savings. Each specification was reviewed by two participants, yielding 120 assessments.
Results: For Relevance, 75% of ratings were highest and 20% second-highest. Clarity reached 90% highest. Completeness: 75% highest, 19% second. Singularity: 82% highest, 12% second. Time Savings: 68% highest, 24% second. No lowest ratings occurred. Mann-Whitney U tests showed no significant differences across participants or models. Llama slightly outperformed Claude in Clarity, Completeness, and Time Savings, while Claude was stronger in Singularity. Feedback noted hallucinations and omissions but confirmed the utility of the specifications.
Conclusion: LLMs can generate high-quality Gherkin specifications from legal texts, reducing manual effort and providing structured artifacts useful for implementation, assurance, and test generation.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:04:34 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Hassani",
"Shabnam",
""
],
[
"Sabetzadeh",
"Mehrdad",
""
],
[
"Amyot",
"Daniel",
""
]
] |
2508.20745
|
Kaustubh Atey
|
Kaustubh Atey, Sameer Anand Jha, Gouranga Bala, Amit Sethi
|
Mix, Align, Distil: Reliable Cross-Domain Atypical Mitosis Classification
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Atypical mitotic figures (AMFs) are important histopathological markers yet remain challenging to identify consistently, particularly under domain shift stemming from scanner, stain, and acquisition differences. We present a simple training-time recipe for domain-robust AMF classification in MIDOG 2025 Task 2. The approach (i) increases feature diversity via style perturbations inserted at early and mid backbone stages, (ii) aligns attention-refined features across sites using weak domain labels (Scanner, Origin, Species, Tumor) through an auxiliary alignment loss, and (iii) stabilizes predictions by distilling from an exponential moving average (EMA) teacher with temperature-scaled KL divergence. On the organizer-run preliminary leaderboard for atypical mitosis classification, our submission attains balanced accuracy of 0.8762, sensitivity of 0.8873, specificity of 0.8651, and ROC AUC of 0.9499. The method incurs negligible inference-time overhead, relies only on coarse domain metadata, and delivers strong, balanced performance, positioning it as a competitive submission for the MIDOG 2025 challenge.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:04:55 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Atey",
"Kaustubh",
""
],
[
"Jha",
"Sameer Anand",
""
],
[
"Bala",
"Gouranga",
""
],
[
"Sethi",
"Amit",
""
]
] |
2508.20746
|
Felipe Marceca
|
Jeremiah Buckley, Felipe Marceca and Joaqu\'in Singer
|
Sampling properties of the zeroes of the Gaussian entire function
| null | null | null | null |
math.PR math.FA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study sampling properties of the zero set of the Gaussian entire function on Fock spaces. Firstly, we relax Seip and Wallst\'en's density and separation conditions for sampling sets on Fock spaces to obtain weighted inequalities for sets that are not necessarily sampling. On the probabilistic front, we estimate the number of zeroes of the Gaussian entire functions that are close to each other. We use these to prove random sampling inequalities for polynomials of degree at most $d$ using ${d}+o(d)$ points, and show that, with high probability, the sampling constants grow slower than $d^\varepsilon$ for any $\varepsilon>0$. In particular, we recover a result from Lyons and Zhai in the case of the Gaussian entire function, where it is shown that the zeroes are (almost surely) a uniqueness set for the Fock space.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:06:43 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Buckley",
"Jeremiah",
""
],
[
"Marceca",
"Felipe",
""
],
[
"Singer",
"Joaquín",
""
]
] |
2508.20747
|
Giovanni Colavizza
|
Giovanni Colavizza, Lauren Cadwallader, Iain Hrynaszkiewicz
|
An analysis of the effects of open science indicators on citations in the French Open Science Monitor
| null | null | null | null |
cs.DL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This study investigates the correlation of citation impact with various open science indicators (OSI) within the French Open Science Monitor (FOSM), a dataset comprising approximately 900,000 publications authored by French authors from 2020 to 2022. By integrating data from OpenAlex and Crossref, we analyze open science indicators such as the presence of a pre-print, data sharing, and software sharing in 576,537 publications in the FOSM dataset. Our analysis reveals a positive correlation between these OSI and citation counts. Considering our most complete citation prediction model, we find pre-prints are correlated with a significant positive effect of 19% on citation counts, software sharing of 13.5%, and data sharing of 14.3%. We find large variations in the correlations of OSIs with citations in different research disciplines, and observe that open access status of publications is correlated with a 8.6% increase in citations in our model. While these results remain observational and are limited to the scope of the analysis, they suggest a consistent correlation between citation advantages and open science indicators. Our results may be valuable to policy makers, funding agencies, researchers, publishers, institutions, and other stakeholders who are interested in understanding the academic impacts, or effects, of open science practices.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:07:50 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Colavizza",
"Giovanni",
""
],
[
"Cadwallader",
"Lauren",
""
],
[
"Hrynaszkiewicz",
"Iain",
""
]
] |
2508.20748
|
Jun Xie
|
Jun Xie, Yuan-hua Ni, Yiqin Yang, Feng Duan, Bo Xu
|
An Efficient Data-Driven Framework for Linear Quadratic Output Feedback Control
| null | null | null | null |
math.OC cs.SY eess.SY
|
http://creativecommons.org/licenses/by/4.0/
|
Linear quadratic regulator with unmeasurable states and unknown system matrix parameters better aligns with practical scenarios. However, for this problem, balancing the optimality of the resulting controller and the leniency of the algorithm's feasibility conditions remains a non-trivial challenge, as no well-established general method has yet been developed to address this trade-off. To address this gap, this study first develops a comprehensive theoretical framework for state parameterization that equivalently substitutes for unknown states. By analyzing the controllability of consistent systems satisfied by substitute states, this framework quantifies the capability of substitute state data matrices to parameterize unknown closed-loop systems and output feedback controllers, thereby constructing a modified state parameterization form that meets the complete data parameterization condition of Willems' Fundamental Lemma. Leveraging this framework, this study proposes efficient model-free off-policy policy iteration and value iteration algorithms with theoretical guarantees to solve for the optimal output feedback controller. Compared with existing studies, particularly for multi-output problems where existing model-free reinforcement learning algorithms may fail, the proposed method removes redundant information in substitute states and the additional full row rank condition on regression matrices, thereby ensuring the solution of optimal output feedback controllers equivalent to optimal state feedback controllers for multi-output systems. Furthermore, this study pioneers a comprehensive and highly scalable theoretical analysis of state parameterization from a data-driven viewpoint, and the proposed algorithms exhibit significant advantages in implementation conditions, data demand, unknown handling, and convergence speed.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:07:59 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Xie",
"Jun",
""
],
[
"Ni",
"Yuan-hua",
""
],
[
"Yang",
"Yiqin",
""
],
[
"Duan",
"Feng",
""
],
[
"Xu",
"Bo",
""
]
] |
2508.20749
|
Fraser Daly
|
Fraser Daly and Andrew Wade
|
Rates of convergence for extremal spacings in Kakutani's random interval-splitting process
|
35 pages
| null | null | null |
math.PR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Kakutani's random interval-splitting process iteratively divides, via a uniformly random splitting point, the largest sub-interval in a partition of the unit interval. The length of the longest sub-interval after $n$ steps, suitably centred and scaled, is known to satisfy a central limit theorem as $n \to \infty$. We provide a quantitative (Berry-Esseen) upper bound for the finite-$n$ approximation in the central limit theorem, with conjecturally optimal rates in $n$. We also prove convergence to an exponential distribution for the length of the smallest sub-interval, with quantitative bounds. The Kakutani process can be embedded in certain branching and fragmentation processes, and we translate our results into that context also. Our proof uses conditioning on an intermediate time, a conditional independence structure for statistics involving small sub-intervals, an Hermite-Edgeworth expansion, and moments estimates with quantitative error bounds.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:08:32 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Daly",
"Fraser",
""
],
[
"Wade",
"Andrew",
""
]
] |
2508.20750
|
Alina Elena Baia
|
Vassiliy Cheremetiev, Quang Long Ho Ngo, Chau Ying Kot, Alina Elena Baia, Andrea Cavallaro
|
Specializing General-purpose LLM Embeddings for Implicit Hate Speech Detection across Datasets
|
Paper accepted at the DHOW Workshop at ACM Multimedia 2025. Code available at https://github.com/idiap/implicit-hsd
| null |
10.1145/3746275.3762209
| null |
cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Implicit hate speech (IHS) is indirect language that conveys prejudice or hatred through subtle cues, sarcasm or coded terminology. IHS is challenging to detect as it does not include explicit derogatory or inflammatory words. To address this challenge, task-specific pipelines can be complemented with external knowledge or additional information such as context, emotions and sentiment data. In this paper, we show that, by solely fine-tuning recent general-purpose embedding models based on large language models (LLMs), such as Stella, Jasper, NV-Embed and E5, we achieve state-of-the-art performance. Experiments on multiple IHS datasets show up to 1.10 percentage points improvements for in-dataset, and up to 20.35 percentage points improvements in cross-dataset evaluation, in terms of F1-macro score.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:08:57 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Cheremetiev",
"Vassiliy",
""
],
[
"Ngo",
"Quang Long Ho",
""
],
[
"Kot",
"Chau Ying",
""
],
[
"Baia",
"Alina Elena",
""
],
[
"Cavallaro",
"Andrea",
""
]
] |
2508.20751
|
Yibin Wang
|
Yibin Wang and Zhimin Li and Yuhang Zang and Yujie Zhou and Jiazi Bu and Chunyu Wang and Qinglin Lu and Cheng Jin and Jiaqi Wang
|
Pref-GRPO: Pairwise Preference Reward-based GRPO for Stable Text-to-Image Reinforcement Learning
|
Project Page: https://codegoat24.github.io/UnifiedReward/Pref-GRPO
| null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Recent advancements highlight the importance of GRPO-based reinforcement learning methods and benchmarking in enhancing text-to-image (T2I) generation. However, current methods using pointwise reward models (RM) for scoring generated images are susceptible to reward hacking. We reveal that this happens when minimal score differences between images are amplified after normalization, creating illusory advantages that drive the model to over-optimize for trivial gains, ultimately destabilizing the image generation process. To address this, we propose Pref-GRPO, a pairwise preference reward-based GRPO method that shifts the optimization objective from score maximization to preference fitting, ensuring more stable training. In Pref-GRPO, images are pairwise compared within each group using preference RM, and the win rate is used as the reward signal. Extensive experiments demonstrate that PREF-GRPO differentiates subtle image quality differences, providing more stable advantages and mitigating reward hacking. Additionally, existing T2I benchmarks are limited by coarse evaluation criteria, hindering comprehensive model assessment. To solve this, we introduce UniGenBench, a unified T2I benchmark comprising 600 prompts across 5 main themes and 20 subthemes. It evaluates semantic consistency through 10 primary and 27 sub-criteria, leveraging MLLM for benchmark construction and evaluation. Our benchmarks uncover the strengths and weaknesses of both open and closed-source T2I models and validate the effectiveness of Pref-GRPO.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:11:24 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Wang",
"Yibin",
""
],
[
"Li",
"Zhimin",
""
],
[
"Zang",
"Yuhang",
""
],
[
"Zhou",
"Yujie",
""
],
[
"Bu",
"Jiazi",
""
],
[
"Wang",
"Chunyu",
""
],
[
"Lu",
"Qinglin",
""
],
[
"Jin",
"Cheng",
""
],
[
"Wang",
"Jiaqi",
""
]
] |
2508.20752
|
Marvin Richter
|
Marvin Richter, Ingrid Strandberg, Simone Gasparinetti, Anton Frisk Kockum
|
Overhead in Quantum Circuits with Time-Multiplexed Qubit Control
| null | null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
When scaling up quantum processors in a cryogenic environment, it is desirable to limit the number of qubit drive lines going into the cryostat, since fewer lines makes cooling of the system more manageable and the need for complicated electronics setups is reduced. However, although time-multiplexing of qubit control enables using just a few drive line to steer many qubits, it comes with a trade-off: fewer drive lines means fewer qubits can be controlled in parallel, which leads to an overhead in the execution time for quantum algorithms. In this article, we quantify this trade-off through numerical and analytical investigations. For standard quantum processor layouts and typical gate times, we show that the trade-off is favorable for many common quantum algorithms $\unicode{x2014}$ the number of drive lines can be significantly reduced without introducing much overhead. Specifically, we show that couplers for two-qubit gates can be grouped on common drive lines without any overhead up to a limit set by the connectivity of the qubits. For single-qubit gates, we find that the serialization overhead generally scales only logarithmically in the number of qubits sharing a drive line. These results are promising for the continued progress towards large-scale quantum computers.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:12:06 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Richter",
"Marvin",
""
],
[
"Strandberg",
"Ingrid",
""
],
[
"Gasparinetti",
"Simone",
""
],
[
"Kockum",
"Anton Frisk",
""
]
] |
2508.20753
|
Hugo Kruiniger Dr
|
Hugo Kruiniger
|
A further look at Modified ML estimation of the panel AR(1) model with fixed effects and arbitrary initial conditions
|
51 pages
| null | null | null |
econ.EM
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this paper we consider two kinds of generalizations of Lancaster's (Review of Economic Studies, 2002) Modified ML estimator (MMLE) for the panel AR(1) model with fixed effects and arbitrary initial conditions and possibly covariates when the time dimension, T, is fixed. When the autoregressive parameter rho=1, the limiting modified profile log-likelihood function for this model has a stationary point of inflection, and rho is first-order underidentified but second-order identified. We show that the generalized MMLEs exist w.p.a.1. and are uniquely defined w.p.1. and consistent for any value of |rho| =< 1. When rho=1, the rate of convergence of the MMLEs is N^{1/4}, where N is the cross-sectional dimension of the panel. We then develop an asymptotic theory for GMM estimators when one of the parameters is only second-order identified and use this to derive the limiting distributions of the MMLEs. They are generally asymmetric when rho=1. We also show that Quasi LM tests that are based on the modified profile log-likelihood and use its expected rather than observed Hessian, with an additional modification for rho=1, and confidence regions based on inverting these tests have correct asymptotic size in a uniform sense when |rho| =< 1. Finally, we investigate the finite sample properties of the MMLEs and the QLM test in a Monte Carlo study.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:12:07 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Kruiniger",
"Hugo",
""
]
] |
2508.20754
|
Yuxi Hu
|
Yuxi Hu, Jun Zhang, Kuangyi Chen, Zhe Zhang, Friedrich Fraundorfer
|
${C}^{3}$-GS: Learning Context-aware, Cross-dimension, Cross-scale Feature for Generalizable Gaussian Splatting
|
Accepted to The 36th British Machine Vision Conference (BMVC 2025), Sheffield, UK
| null | null | null |
cs.CV cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Generalizable Gaussian Splatting aims to synthesize novel views for unseen scenes without per-scene optimization. In particular, recent advancements utilize feed-forward networks to predict per-pixel Gaussian parameters, enabling high-quality synthesis from sparse input views. However, existing approaches fall short in encoding discriminative, multi-view consistent features for Gaussian predictions, which struggle to construct accurate geometry with sparse views. To address this, we propose $\mathbf{C}^{3}$-GS, a framework that enhances feature learning by incorporating context-aware, cross-dimension, and cross-scale constraints. Our architecture integrates three lightweight modules into a unified rendering pipeline, improving feature fusion and enabling photorealistic synthesis without requiring additional supervision. Extensive experiments on benchmark datasets validate that $\mathbf{C}^{3}$-GS achieves state-of-the-art rendering quality and generalization ability. Code is available at: https://github.com/YuhsiHu/C3-GS.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:12:18 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Hu",
"Yuxi",
""
],
[
"Zhang",
"Jun",
""
],
[
"Chen",
"Kuangyi",
""
],
[
"Zhang",
"Zhe",
""
],
[
"Fraundorfer",
"Friedrich",
""
]
] |
2508.20755
|
Ambroise Odonnat
|
Sam Houliston and Ambroise Odonnat and Charles Arnal and Vivien Cabannes
|
Provable Benefits of In-Tool Learning for Large Language Models
| null | null | null | null |
cs.LG cs.AI stat.ML
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Tool-augmented language models, equipped with retrieval, memory, or external APIs, are reshaping AI, yet their theoretical advantages remain underexplored. In this paper, we address this question by demonstrating the benefits of in-tool learning (external retrieval) over in-weight learning (memorization) for factual recall. We show that the number of facts a model can memorize solely in its weights is fundamentally limited by its parameter count. In contrast, we prove that tool-use enables unbounded factual recall via a simple and efficient circuit construction. These results are validated in controlled experiments, where tool-using models consistently outperform memorizing ones. We further show that for pretrained large language models, teaching tool-use and general rules is more effective than finetuning facts into memory. Our work provides both a theoretical and empirical foundation, establishing why tool-augmented workflows are not just practical, but provably more scalable.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:12:19 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Houliston",
"Sam",
""
],
[
"Odonnat",
"Ambroise",
""
],
[
"Arnal",
"Charles",
""
],
[
"Cabannes",
"Vivien",
""
]
] |
2508.20756
|
Jacopo Romano
|
Jacopo Romano and Andrea Gambassi
|
Anomalous diffusion and run-and-tumble motion of a chemotactic particle in low dimensions
| null | null | null | null |
cond-mat.stat-mech
|
http://creativecommons.org/licenses/by/4.0/
|
We study the stochastic dynamics of a symmetric self-chemotactic particle and determine the long-time behavior of its mean squared displacement (MSD). The attractive or repulsive interaction of the particle with the chemical field that it generates induces a non-linear, non-Markovian effective dynamics, which results into anomalous diffusion for spatial dimensions $d \leq 2$. In one spatial dimension, we map the case of repulsive chemotaxis onto a run-and-tumble-like dynamics, leading to an MSD which, as a function of the elapsed time $t$, grows superdiffusively with exponent $4/3$. In the presence of attractive chemotaxis, instead, the particle exhibits a slowdown, with the MSD growing logarithmically with time. In $d=2$, we find logarithmic aging of the diffusion coefficient, while in $d=3$ the motion reverts standard diffusive behavior with a renormalized diffusion coefficient.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:13:35 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Romano",
"Jacopo",
""
],
[
"Gambassi",
"Andrea",
""
]
] |
2508.20757
|
Yuanhao Ding
|
Yuanhao Ding and Esteban Garces Arias and Meimingwei Li and Julian Rodemann and Matthias A{\ss}enmacher and Danlu Chen and Gaojuan Fan and Christian Heumann and Chongsheng Zhang
|
GUARD: Glocal Uncertainty-Aware Robust Decoding for Effective and Efficient Open-Ended Text Generation
|
Accepted at Findings of the Association for Computational Linguistics: EMNLP (Findings) 2025
| null | null | null |
cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Open-ended text generation faces a critical challenge: balancing coherence with diversity in LLM outputs. While contrastive search-based decoding strategies have emerged to address this trade-off, their practical utility is often limited by hyperparameter dependence and high computational costs. We introduce GUARD, a self-adaptive decoding method that effectively balances these competing objectives through a novel "Glocal" uncertainty-driven framework. GUARD combines global entropy estimates with local entropy deviations to integrate both long-term and short-term uncertainty signals. We demonstrate that our proposed global entropy formulation effectively mitigates abrupt variations in uncertainty, such as sudden overconfidence or high entropy spikes, and provides theoretical guarantees of unbiasedness and consistency. To reduce computational overhead, we incorporate a simple yet effective token-count-based penalty into GUARD. Experimental results demonstrate that GUARD achieves a good balance between text diversity and coherence, while exhibiting substantial improvements in generation speed. In a more nuanced comparison study across different dimensions of text quality, both human and LLM evaluators validated its remarkable performance. Our code is available at https://github.com/YecanLee/GUARD.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:14:20 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Ding",
"Yuanhao",
""
],
[
"Arias",
"Esteban Garces",
""
],
[
"Li",
"Meimingwei",
""
],
[
"Rodemann",
"Julian",
""
],
[
"Aßenmacher",
"Matthias",
""
],
[
"Chen",
"Danlu",
""
],
[
"Fan",
"Gaojuan",
""
],
[
"Heumann",
"Christian",
""
],
[
"Zhang",
"Chongsheng",
""
]
] |
2508.20758
|
Jiawen Lin
|
Jiawen Lin, Shiran Bian, Yihang Zhu, Wenbin Tan, Yachao Zhang, Yuan Xie, Yanyun Qu
|
SeqVLM: Proposal-Guided Multi-View Sequences Reasoning via VLM for Zero-Shot 3D Visual Grounding
| null | null | null | null |
cs.CV cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
3D Visual Grounding (3DVG) aims to localize objects in 3D scenes using natural language descriptions. Although supervised methods achieve higher accuracy in constrained settings, zero-shot 3DVG holds greater promise for real-world applications since eliminating scene-specific training requirements. However, existing zero-shot methods face challenges of spatial-limited reasoning due to reliance on single-view localization, and contextual omissions or detail degradation. To address these issues, we propose SeqVLM, a novel zero-shot 3DVG framework that leverages multi-view real-world scene images with spatial information for target object reasoning. Specifically, SeqVLM first generates 3D instance proposals via a 3D semantic segmentation network and refines them through semantic filtering, retaining only semantic-relevant candidates. A proposal-guided multi-view projection strategy then projects these candidate proposals onto real scene image sequences, preserving spatial relationships and contextual details in the conversion process of 3D point cloud to images. Furthermore, to mitigate VLM computational overload, we implement a dynamic scheduling mechanism that iteratively processes sequances-query prompts, leveraging VLM's cross-modal reasoning capabilities to identify textually specified objects. Experiments on the ScanRefer and Nr3D benchmarks demonstrate state-of-the-art performance, achieving Acc@0.25 scores of 55.6% and 53.2%, surpassing previous zero-shot methods by 4.0% and 5.2%, respectively, which advance 3DVG toward greater generalization and real-world applicability. The code is available at https://github.com/JiawLin/SeqVLM.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:15:37 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Lin",
"Jiawen",
""
],
[
"Bian",
"Shiran",
""
],
[
"Zhu",
"Yihang",
""
],
[
"Tan",
"Wenbin",
""
],
[
"Zhang",
"Yachao",
""
],
[
"Xie",
"Yuan",
""
],
[
"Qu",
"Yanyun",
""
]
] |
2508.20759
|
Kaixuan Huang
|
Ziting Wang, Zi-Yong Ge, Yun-Hao Shi, Zheng-An Wang, Si-Yun Zhou, Hao Li, Kui Zhao, Yue-Shan Xu, Wei-Guo Ma, Hao-Tian Liu, Cai-Ping Fang, Jia-Cheng Song, Tian-Ming Li, Jia-Chi Zhang, Yu Liu, Cheng-Lin Deng, Guangming Xue, Haifeng Yu, Kai Xu, Kaixuan Huang, Franco Nori, Heng Fan
|
Observation of Inelastic Meson Scattering in a Floquet System using a Digital Quantum Simulator
|
7+3 pages, 4+2 figures
| null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Lattice gauge theories provide a non-perturbative framework for understanding confinement and hadronic physics, but their real-time dynamics remain challenging for classical computations. However, quantum simulators offer a promising alternative for exploring such dynamics beyond classical capabilities. Here, we experimentally investigate meson scattering using a superconducting quantum processor. Employing a digital protocol, we realize a Floquet spin chain equivalent to a one-dimensional Floquet $\mathbb{Z}_2$ lattice gauge theory. We observe Bloch oscillations of single kinks and strong binding between adjacent kinks, signaling confinement and the formation of stable mesons in this Floquet system. Using full-system joint readout, we resolve meson populations by string length, enabling identification of meson scattering channels. Our results reveal the fragmentation of a long-string meson into multiple short-string mesons, which is also an experimental signature of string breaking. Moreover, we directly observe inelastic meson scattering, where two short-string mesons can merge into a longer one. Our results pave the way for studying interacting gauge particles and composite excitations on digital quantum simulators.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:16:39 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Wang",
"Ziting",
""
],
[
"Ge",
"Zi-Yong",
""
],
[
"Shi",
"Yun-Hao",
""
],
[
"Wang",
"Zheng-An",
""
],
[
"Zhou",
"Si-Yun",
""
],
[
"Li",
"Hao",
""
],
[
"Zhao",
"Kui",
""
],
[
"Xu",
"Yue-Shan",
""
],
[
"Ma",
"Wei-Guo",
""
],
[
"Liu",
"Hao-Tian",
""
],
[
"Fang",
"Cai-Ping",
""
],
[
"Song",
"Jia-Cheng",
""
],
[
"Li",
"Tian-Ming",
""
],
[
"Zhang",
"Jia-Chi",
""
],
[
"Liu",
"Yu",
""
],
[
"Deng",
"Cheng-Lin",
""
],
[
"Xue",
"Guangming",
""
],
[
"Yu",
"Haifeng",
""
],
[
"Xu",
"Kai",
""
],
[
"Huang",
"Kaixuan",
""
],
[
"Nori",
"Franco",
""
],
[
"Fan",
"Heng",
""
]
] |
2508.20760
|
Alma Liezenga
|
Jan Erik van Woerden, Gertjan Burghouts, Lotte Nijskens, Alma M. Liezenga, Sabina van Rooij, Frank Ruis, Hugo J. Kuijf
|
Occlusion Robustness of CLIP for Military Vehicle Classification
|
To be presented at SPIE: Sensors + Imaging, Artificial Intelligence for Security and Defence Applications II
| null | null | null |
cs.CV cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Vision-language models (VLMs) like CLIP enable zero-shot classification by aligning images and text in a shared embedding space, offering advantages for defense applications with scarce labeled data. However, CLIP's robustness in challenging military environments, with partial occlusion and degraded signal-to-noise ratio (SNR), remains underexplored. We investigate CLIP variants' robustness to occlusion using a custom dataset of 18 military vehicle classes and evaluate using Normalized Area Under the Curve (NAUC) across occlusion percentages. Four key insights emerge: (1) Transformer-based CLIP models consistently outperform CNNs, (2) fine-grained, dispersed occlusions degrade performance more than larger contiguous occlusions, (3) despite improved accuracy, performance of linear-probed models sharply drops at around 35% occlusion, (4) by finetuning the model's backbone, this performance drop occurs at more than 60% occlusion. These results underscore the importance of occlusion-specific augmentations during training and the need for further exploration into patch-level sensitivity and architectural resilience for real-world deployment of CLIP.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:16:55 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"van Woerden",
"Jan Erik",
""
],
[
"Burghouts",
"Gertjan",
""
],
[
"Nijskens",
"Lotte",
""
],
[
"Liezenga",
"Alma M.",
""
],
[
"van Rooij",
"Sabina",
""
],
[
"Ruis",
"Frank",
""
],
[
"Kuijf",
"Hugo J.",
""
]
] |
2508.20761
|
Tirza Routtenberg
|
Yaniv Mazor and Tirza Routtenberg
|
Weighted Bayesian Cram$\acute{\text{e}}$r-Rao Bound for Mixed-Resolution Parameter Estimation
| null | null | null | null |
eess.SP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Mixed-resolution architectures, combining high-resolution (analog) data with coarsely quantized (e.g., 1-bit) data, are widely employed in emerging communication and radar systems to reduce hardware costs and power consumption. However, the use of coarsely quantized data introduces non-trivial tradeoffs in parameter estimation tasks. In this paper, we investigate the derivation of lower bounds for such systems. In particular, we develop the weighted Bayesian Cramer-Rao bound (WBCRB) for the mixed-resolution setting with a general weight function. We demonstrate the special cases of: (i) the classical BCRB; (ii) the WBCRB that is based on the Bayesian Fisher information matrix (BFIM)-Inverse weighting; and (iii) the Aharon-Tabrikian tightest WBCRB with an optimal weight function. Based on the developed WBCRB, we propose a new method to approximate the mean-squared-error (MSE) by partitioning the estimation problem into two regions: (a) where the 1-bit quantized data is informative; and (b) where it is saturated. We apply region-specific WBCRB approximations in these regions to achieve an accurate composite MSE estimate. We derive the bounds and MSE approximation for the linear Gaussian orthonormal (LGO) model, which is commonly used in practical signal processing applications. Our simulation results demonstrate the use of the proposed bounds and approximation method in the LGO model with a scalar unknown parameter. It is shown that the WBCRB outperforms the BCRB, where the BFIM-Inverse weighting version approaches the optimal WBCRB. Moreover, it is shown that the WBCRB-based MSE approximation is tighter and accurately predicts the non-monotonic behavior of the MSE in the presence of quantization errors.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:17:35 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Mazor",
"Yaniv",
""
],
[
"Routtenberg",
"Tirza",
""
]
] |
2508.20762
|
Fachri Najm Noer Kartiman Mr
|
Fachri Najm Noer Kartiman, Rasim, Yaya Wihardi, Nurul Hasanah, Oskar Natan, Bambang Wahono, Taufik Ibnu Salim
|
SKGE-SWIN: End-To-End Autonomous Vehicle Waypoint Prediction and Navigation Using Skip Stage Swin Transformer
|
keywords-multitask learning, autonomous driving, end-to-end learning, skip connections, swin transformer, self-attention mechanism. 12 pages
| null | null | null |
cs.CV cs.AI cs.LG cs.RO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Focusing on the development of an end-to-end autonomous vehicle model with pixel-to-pixel context awareness, this research proposes the SKGE-Swin architecture. This architecture utilizes the Swin Transformer with a skip-stage mechanism to broaden feature representation globally and at various network levels. This approach enables the model to extract information from distant pixels by leveraging the Swin Transformer's Shifted Window-based Multi-head Self-Attention (SW-MSA) mechanism and to retain critical information from the initial to the final stages of feature extraction, thereby enhancing its capability to comprehend complex patterns in the vehicle's surroundings. The model is evaluated on the CARLA platform using adversarial scenarios to simulate real-world conditions. Experimental results demonstrate that the SKGE-Swin architecture achieves a superior Driving Score compared to previous methods. Furthermore, an ablation study will be conducted to evaluate the contribution of each architectural component, including the influence of skip connections and the use of the Swin Transformer, in improving model performance.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:17:35 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Kartiman",
"Fachri Najm Noer",
""
],
[
"Rasim",
"",
""
],
[
"Wihardi",
"Yaya",
""
],
[
"Hasanah",
"Nurul",
""
],
[
"Natan",
"Oskar",
""
],
[
"Wahono",
"Bambang",
""
],
[
"Salim",
"Taufik Ibnu",
""
]
] |
2508.20763
|
Rachel Steinhorst
|
Aleksi Kurkela and Krishna Rajagopal and Rachel Steinhorst
|
Updated Astrophysical Equation-of-State Constraints on the Color-Superconducting Gap
|
5 pages, 2 figures, proceedings of Quark Matter 2025
| null | null |
MIT-CTP/5909
|
hep-ph astro-ph.HE nucl-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We summarize and update using new NICER measurements the results of arXiv:2401.16253, in which we used various astrophysical neutron-star observations to set an upper bound on the CFL color-superconducting gap in a range of baryon chemical potentials $\mu_B \in [2.1,3.2]$, above those reached within neutron stars. We also corroborate the ``reasonable" constraint from arXiv:2401.16253 on the maximum value of the color-superconducting gap by performing a new Bayesian analysis using a prior that extends a two-segment Gaussian process connecting the whole density range between CEFT and pQCD.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:17:59 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Kurkela",
"Aleksi",
""
],
[
"Rajagopal",
"Krishna",
""
],
[
"Steinhorst",
"Rachel",
""
]
] |
2508.20764
|
Xiaoyi Wang
|
Xiaoyi Wang, Jiwei Zhang, Guangtao Zhang, Honglei Guo
|
Feel the Difference? A Comparative Analysis of Emotional Arcs in Real and LLM-Generated CBT Sessions
|
Accepted at EMNLP 2025,14 page,3 figures
| null | null | null |
cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Synthetic therapy dialogues generated by large language models (LLMs) are increasingly used in mental health NLP to simulate counseling scenarios, train models, and supplement limited real-world data. However, it remains unclear whether these synthetic conversations capture the nuanced emotional dynamics of real therapy. In this work, we conduct the first comparative analysis of emotional arcs between real and LLM-generated Cognitive Behavioral Therapy dialogues. We adapt the Utterance Emotion Dynamics framework to analyze fine-grained affective trajectories across valence, arousal, and dominance dimensions. Our analysis spans both full dialogues and individual speaker roles (counselor and client), using real sessions transcribed from public videos and synthetic dialogues from the CACTUS dataset. We find that while synthetic dialogues are fluent and structurally coherent, they diverge from real conversations in key emotional properties: real sessions exhibit greater emotional variability,more emotion-laden language, and more authentic patterns of reactivity and regulation. Moreover, emotional arc similarity between real and synthetic speakers is low, especially for clients. These findings underscore the limitations of current LLM-generated therapy data and highlight the importance of emotional fidelity in mental health applications. We introduce RealCBT, a curated dataset of real CBT sessions, to support future research in this space.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:19:31 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Wang",
"Xiaoyi",
""
],
[
"Zhang",
"Jiwei",
""
],
[
"Zhang",
"Guangtao",
""
],
[
"Guo",
"Honglei",
""
]
] |
2508.20765
|
Gowreesh Mago
|
Gowreesh Mago, Pascal Mettes, Stevan Rudinac
|
Looking Beyond the Obvious: A Survey on Abstract Concept Recognition for Video Understanding
|
Under Review for IJCV
| null | null | null |
cs.CV cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
The automatic understanding of video content is advancing rapidly. Empowered by deeper neural networks and large datasets, machines are increasingly capable of understanding what is concretely visible in video frames, whether it be objects, actions, events, or scenes. In comparison, humans retain a unique ability to also look beyond concrete entities and recognize abstract concepts like justice, freedom, and togetherness. Abstract concept recognition forms a crucial open challenge in video understanding, where reasoning on multiple semantic levels based on contextual information is key. In this paper, we argue that the recent advances in foundation models make for an ideal setting to address abstract concept understanding in videos. Automated understanding of high-level abstract concepts is imperative as it enables models to be more aligned with human reasoning and values. In this survey, we study different tasks and datasets used to understand abstract concepts in video content. We observe that, periodically and over a long period, researchers have attempted to solve these tasks, making the best use of the tools available at their disposal. We advocate that drawing on decades of community experience will help us shed light on this important open grand challenge and avoid ``re-inventing the wheel'' as we start revisiting it in the era of multi-modal foundation models.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:19:49 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Mago",
"Gowreesh",
""
],
[
"Mettes",
"Pascal",
""
],
[
"Rudinac",
"Stevan",
""
]
] |
2508.20766
|
Hasan Abed Al Kader Hammoud
|
Harethah Abu Shairah, Hasan Abed Al Kader Hammoud, George Turkiyyah, Bernard Ghanem
|
Turning the Spell Around: Lightweight Alignment Amplification via Rank-One Safety Injection
|
Under Review
| null | null | null |
cs.CL cs.AI cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Safety alignment in Large Language Models (LLMs) often involves mediating internal representations to refuse harmful requests. Recent research has demonstrated that these safety mechanisms can be bypassed by ablating or removing specific representational directions within the model. In this paper, we propose the opposite approach: Rank-One Safety Injection (ROSI), a white-box method that amplifies a model's safety alignment by permanently steering its activations toward the refusal-mediating subspace. ROSI operates as a simple, fine-tuning-free rank-one weight modification applied to all residual stream write matrices. The required safety direction can be computed from a small set of harmful and harmless instruction pairs. We show that ROSI consistently increases safety refusal rates - as evaluated by Llama Guard 3 - while preserving the utility of the model on standard benchmarks such as MMLU, HellaSwag, and Arc. Furthermore, we show that ROSI can also re-align 'uncensored' models by amplifying their own latent safety directions, demonstrating its utility as an effective last-mile safety procedure. Our results suggest that targeted, interpretable weight steering is a cheap and potent mechanism to improve LLM safety, complementing more resource-intensive fine-tuning paradigms.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:22:33 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Shairah",
"Harethah Abu",
""
],
[
"Hammoud",
"Hasan Abed Al Kader",
""
],
[
"Turkiyyah",
"George",
""
],
[
"Ghanem",
"Bernard",
""
]
] |
2508.20767
|
Teresa Karolin Bister
|
Teresa Bister
|
Probing UHECR sources - Constraints from cosmic-ray measurements
|
24 pages, Review presented at the 39th ICRC conference in Geneva (PoS ICRC 2025 1387)
| null | null | null |
astro-ph.HE
|
http://creativecommons.org/licenses/by-sa/4.0/
|
Ultra-high-energy cosmic rays (UHECRs) are the most energetic particles known - and yet their origin is still an open question. However, with the precision and accumulated statistics of the Pierre Auger Observatory and the Telescope Array, in combination with advancements in theory and modeling - e.g. of the Galactic magnetic field - it is now possible to set solid constraints on the sources of UHECRs. The spectrum and composition measurements above the ankle can be well described by a population of extragalactic, homogeneously distributed sources emitting mostly intermediate-mass nuclei. Using additionally the observed anisotropy in the arrival directions, namely the large-scale dipole >8 EeV as well as smaller-scale warmspots at higher energies, even more powerful constraints on the density and distribution of sources can be placed. Yet, open questions remain - like the striking similarity of the sources that is necessary to describe the rather pure mass composition above the ankle, or the origin of the highest energy events whose tracked back directions point towards voids. The current findings and possible interpretation of UHECR data will be presented in this review.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:23:24 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Bister",
"Teresa",
""
]
] |
2508.20768
|
Isaac Harris
|
Isaac Harris, Andreas Kleefeld, and Heejin Lee
|
Existence of transmission eigenvalues for biharmonic scattering by a clamped planar region
| null | null | null | null |
math.AP math-ph math.MP
|
http://creativecommons.org/licenses/by/4.0/
|
In this paper, we study the so-called clamped transmission eigenvalue problem. This is a new transmission eigenvalue problem that is derived from the scattering of an impenetrable clamped obstacle in a thin elastic plate. The scattering problem is modeled by a biharmonic wave operator given by the Kirchhoff--Love infinite plate problem in the frequency domain. These scattering problems have not been studied to the extent of other models. Unlike other transmission eigenvalue problems, the problem studied here is a system of homogeneous PDEs defined in all of $\mathbb{R}^2$. This provides unique analytical and computational difficulties when studying the clamped transmission eigenvalue problem. We are able to prove that there exist infinitely many real clamped transmission eigenvalues. This is done by studying the equivalent variational formulation. We also investigate the relationship of the clamped transmission eigenvalues to the Dirichlet and Neumann eigenvalues of the negative Laplacian for the bounded scattering obstacle.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:24:16 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Harris",
"Isaac",
""
],
[
"Kleefeld",
"Andreas",
""
],
[
"Lee",
"Heejin",
""
]
] |
2508.20769
|
Ashok Singal
|
Ashok K. Singal
|
Solar peculiar motion inferred from dipole anisotropy in redshift distribution of quasars appears to lie along the Galactic Centre direction
|
16 pages, 6 figures, 1 Table, accepted in Scientific Reports - Nature Portfolio journal
| null | null | null |
astro-ph.CO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
According to the Cosmological Principle an observer stationary with respect to the comoving coordinates of the expanding universe should find the redshift distribution of distant quasars to be isotropic. However, the observed redshift distribution in a large sample of 1.3 million quasars shows a significant dipole anisotropy. A peculiar motion of the observer could introduce such a dipole anisotropy in the observed redshift distribution. However, the motion inferred therefrom turns out to be not only many times the peculiar motion estimated from the anisotropy in the Cosmic Microwave Background (CMB), but also nearly in a direction at a right angle. The Solar peculiar motion, in fact, turns out to be, quite unexpectedly, in the direction of the Galactic Centre. Such a statistically significant discrepancy in peculiar motion, derived by different methodologies, could imply a violation of the cosmological principle, a cornerstone in the foundation of the standard model.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:26:13 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Singal",
"Ashok K.",
""
]
] |
2508.20770
|
Guillermo Romero
|
Francisco Albarr\'an-Arriagada, Guillermo Romero, Juan Carlos Retamal
|
Equally entangled multiqubit states
|
8 pages, 8 figures
| null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present a protocol for generating multiqubit quantum states with translationally invariant pairwise entanglement. Our approach is tailored for digital quantum computers with restricted qubit connectivity, a common limitation in state-of-the-art hardware platforms. We examine two configurations: star connectivity, which enables rotationally invariant entanglement, and linear connectivity, which achieves translationally invariant entanglement. For the linear configuration, we use a variant of the time-dependent density matrix renormalization group (tDMRG) algorithm to demonstrate that our protocol is independent of the qubits' number. A slight modification of the protocol reveals the presence of quantum states that exhibit periodicity of entanglement among nearest-neighbor qubits. The configurations and protocols of this work are well-suited for near-term quantum devices, offering a feasible route for the experimental realization of symmetric entangled states.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:26:13 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Albarrán-Arriagada",
"Francisco",
""
],
[
"Romero",
"Guillermo",
""
],
[
"Retamal",
"Juan Carlos",
""
]
] |
2508.20771
|
Enrico Liscio
|
Abhishek Kuber, Enrico Liscio, Ruixuan Zhang, Caroline Figueroa, Pradeep K. Murukannaiah
|
Signs of Struggle: Spotting Cognitive Distortions across Language and Register
| null | null | null | null |
cs.CL cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Rising mental health issues among youth have increased interest in automated approaches for detecting early signs of psychological distress in digital text. One key focus is the identification of cognitive distortions, irrational thought patterns that have a role in aggravating mental distress. Early detection of these distortions may enable timely, low-cost interventions. While prior work has focused on English clinical data, we present the first in-depth study of cross-lingual and cross-register generalization of cognitive distortion detection, analyzing forum posts written by Dutch adolescents. Our findings show that while changes in language and writing style can significantly affect model performance, domain adaptation methods show the most promise.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:28:07 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Kuber",
"Abhishek",
""
],
[
"Liscio",
"Enrico",
""
],
[
"Zhang",
"Ruixuan",
""
],
[
"Figueroa",
"Caroline",
""
],
[
"Murukannaiah",
"Pradeep K.",
""
]
] |
2508.20772
|
Mikhail Zubkov Dr
|
M.A.Zubkov
|
Topological invariant responsible for the stability of the Fermi surfaces in non - homogeneous systems
|
Latex, 10 pages; Supplementary Materials, 6 pages
| null | null | null |
cond-mat.mes-hall
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The topological invariant responsible for the stability of Fermi point/Fermi surface in homogeneous systems is expressed through the one particle Green function, which depends on momentum. It is given by an integral over the 3D hypersurface in momentum space surrounding the Fermi surface. Notion of Fermi surface may be extended to the non - homogeneous systems using Wigner - Weyl calculus. The Fermi surface becomes coordinate dependent, it may be defined as the position of the singularity in momentum space of the Wigner transformed Green function. Then the topological invariant responsible for the stability of this Fermi surface is given by the same expression as for the homogeneous case, in which the Green function is replaced by its Wigner transformation while the ordinary products are replaced by the Moyal products. We illustrate the proposed construction by the examples of the systems, in which the given topological invariant is nontrivial and may be calculated explicitly.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:28:48 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Zubkov",
"M. A.",
""
]
] |
2508.20773
|
Christoforos Spartalis
|
Christoforos N. Spartalis, Theodoros Semertzidis, Petros Daras, Efstratios Gavves
|
Unleashing Uncertainty: Efficient Machine Unlearning for Generative AI
|
ICML 2025 workshop on Machine Unlearning for Generative AI
| null | null | null |
cs.LG cs.AI cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
We introduce SAFEMax, a novel method for Machine Unlearning in diffusion models. Grounded in information-theoretic principles, SAFEMax maximizes the entropy in generated images, causing the model to generate Gaussian noise when conditioned on impermissible classes by ultimately halting its denoising process. Also, our method controls the balance between forgetting and retention by selectively focusing on the early diffusion steps, where class-specific information is prominent. Our results demonstrate the effectiveness of SAFEMax and highlight its substantial efficiency gains over state-of-the-art methods.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:29:21 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Spartalis",
"Christoforos N.",
""
],
[
"Semertzidis",
"Theodoros",
""
],
[
"Daras",
"Petros",
""
],
[
"Gavves",
"Efstratios",
""
]
] |
2508.20774
|
Markus Funke
|
Markus Funke and Patricia Lago
|
Towards an Architectural Perspective for Sustainability: Bundle the Needs from Industry
| null | null | null | null |
cs.SE
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Sustainability is increasingly recognized as an emerging quality property in software-intensive systems, yet architects lack structured guidance to address it effectively throughout the software design phase. Architectural perspectives-an architectural knowledge artifact composed of concerns, activities, tactics, pitfalls, and checklists-offer a promising approach to tackle such emerging quality properties across architectural views and are also independent of architecture frameworks and industry contexts. In this paper, we present a sustainability perspective vision, i.e., a revised notion of architectural perspective meant to be filled with its own elements to target sustainability concerns. We formulate our sustainability perspective vision through evidence from applying snowballing to seminal literature and from conducting a focus group with experts in the field. Our findings confirm the relevance of the different perspective elements in practice and highlight implications for shaping a sustainability perspective that meets industrial needs.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:30:48 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Funke",
"Markus",
""
],
[
"Lago",
"Patricia",
""
]
] |
2508.20775
|
Lukas Simon
|
Hua-Sheng Shao, Lukas Simon
|
Progress in NLO Calculations for gamma-gamma Physics
|
10 pages, 5 figures, 1 table; 2nd International Workshop on the physics of Ultra Peripheral Collisions (UPC 2025)
| null | null | null |
hep-ph
|
http://creativecommons.org/licenses/by/4.0/
|
With the advent of precision measurements of photon-fusion processes in ultraperipheral collisions (UPCs) at facilities, such as RHIC and the LHC, the inclusion of higher-order corrections has become essential. While automated frameworks already make NLO corrections feasible for parton-parton scattering processes, no comparable tools had previously been available for UPC processes in two-photon collisions. In these proceedings, we review some recent explicit NLO computations and present the updated gamma-UPC+MadGraph5_aMC@NLO framework, the first automated software that enables NLO-accurate predictions for photon-photon processes in UPCs.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:30:52 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Shao",
"Hua-Sheng",
""
],
[
"Simon",
"Lukas",
""
]
] |
2508.20776
|
Kuniko Paxton
|
Kuniko Paxton, Koorosh Aslansefat, Amila Akagi\'c, Dhavalkumar Thakker, Yiannis Papadopoulos
|
Safer Skin Lesion Classification with Global Class Activation Probability Map Evaluation and SafeML
| null | null | null | null |
cs.CV cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Recent advancements in skin lesion classification models have significantly improved accuracy, with some models even surpassing dermatologists' diagnostic performance. However, in medical practice, distrust in AI models remains a challenge. Beyond high accuracy, trustworthy, explainable diagnoses are essential. Existing explainability methods have reliability issues, with LIME-based methods suffering from inconsistency, while CAM-based methods failing to consider all classes. To address these limitations, we propose Global Class Activation Probabilistic Map Evaluation, a method that analyses all classes' activation probability maps probabilistically and at a pixel level. By visualizing the diagnostic process in a unified manner, it helps reduce the risk of misdiagnosis. Furthermore, the application of SafeML enhances the detection of false diagnoses and issues warnings to doctors and patients as needed, improving diagnostic reliability and ultimately patient safety. We evaluated our method using the ISIC datasets with MobileNetV2 and Vision Transformers.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:32:35 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Paxton",
"Kuniko",
""
],
[
"Aslansefat",
"Koorosh",
""
],
[
"Akagić",
"Amila",
""
],
[
"Thakker",
"Dhavalkumar",
""
],
[
"Papadopoulos",
"Yiannis",
""
]
] |
2508.20777
|
Chuan-Qi He
|
Xiang Chen, Xin Guan, Chuan-Qi He, Yan-Qing Ma, Jian Wang and Da-Jiang Zhang
|
Analytical two-loop amplitudes of $e^{+} e^{-} \longrightarrow \boldsymbol{J} / \boldsymbol{\psi}+\boldsymbol{\eta}_c$ at $B$ factories
|
11 pages
| null | null | null |
hep-ph hep-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In double charmonium production, a long-standing challenge is that the theoretical predictions are not consistent with the measurements at B factories. Within the NRQCD framework, the next-to-leading order (NLO) calculation has proved its power to cut down the discrepancy between theory and experiments. To further clarify this puzzle, we have performed the next-to-next-to-leading order (NNLO) calculation. The amplitude is obtained as an analytical asymptotic expansion in the ratio of the squared charm-quark mass over the squared center-of-mass energy, $m_c^2/s$. We investigate the origin of the leading logarithms by performing a region analysis, revealing the intricate factorization structure in this process. We provide numerical predictions on the total cross sections of $J/\psi+\eta_c$ production, which agree with the experimental results. Extension of our computation to $\Upsilon+\eta_b$ production is also discussed.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:33:02 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Chen",
"Xiang",
""
],
[
"Guan",
"Xin",
""
],
[
"He",
"Chuan-Qi",
""
],
[
"Ma",
"Yan-Qing",
""
],
[
"Wang",
"Jian",
""
],
[
"Zhang",
"Da-Jiang",
""
]
] |
2508.20778
|
Xinhao Huang
|
Xinhao Huang, Zhibo Ren, Yipeng Yu, Ying Zhou, Zulong Chen, Zeyi Wen
|
SEAL: Structure and Element Aware Learning to Improve Long Structured Document Retrieval
|
Accepted at EMNLP 2025 Main Conference
| null | null | null |
cs.IR cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In long structured document retrieval, existing methods typically fine-tune pre-trained language models (PLMs) using contrastive learning on datasets lacking explicit structural information. This practice suffers from two critical issues: 1) current methods fail to leverage structural features and element-level semantics effectively, and 2) the lack of datasets containing structural metadata. To bridge these gaps, we propose \our, a novel contrastive learning framework. It leverages structure-aware learning to preserve semantic hierarchies and masked element alignment for fine-grained semantic discrimination. Furthermore, we release \dataset, a long structured document retrieval dataset with rich structural annotations. Extensive experiments on both released and industrial datasets across various modern PLMs, along with online A/B testing, demonstrate consistent performance improvements, boosting NDCG@10 from 73.96\% to 77.84\% on BGE-M3. The resources are available at https://github.com/xinhaoH/SEAL.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:34:42 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Huang",
"Xinhao",
""
],
[
"Ren",
"Zhibo",
""
],
[
"Yu",
"Yipeng",
""
],
[
"Zhou",
"Ying",
""
],
[
"Chen",
"Zulong",
""
],
[
"Wen",
"Zeyi",
""
]
] |
2508.20779
|
Basile Herzog
|
Basile Herzog, Patrik Thunstr\"om, Olle Eriksson
|
A configuration interaction approach to solve the Anderson impurity model; applications to elemental Ce
|
12 pages, 10 figures
| null | null | null |
cond-mat.str-el
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Accurate calculations of strongly correlated materials remain a formidable challenge in condensed matter physics, particularly due to the computational demand of conventional methods. This paper presents an efficient solver for dynamical mean field theory using configuration interaction (CI). The method is shown to have improved efficiency compared to traditional, exact diagonalization approaches. Hence, it provides an accessible, open-source alternative that can be executed on standard laptop computers or on supercomputers. The solver is demonstrated on cerium in the $\gamma$-, $\alpha$- and $\epsilon$-phases. An analysis of how the electronic structure of Ce evolves as function of lattice compression is made. It is argued that the electronic structure evolves from a localized nature of the 4f shell in $\gamma$-Ce to an essentially itinerant nature of the 4f shell of $\epsilon$-Ce. The transition between these two phases, as function of compression, can hence be seen as a Mott transition. However, this transition is intercepted by the strongly correlated $\alpha$-phase of elemental Ce, for which the 4f shell forms a Kondo singlet.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:35:44 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Herzog",
"Basile",
""
],
[
"Thunström",
"Patrik",
""
],
[
"Eriksson",
"Olle",
""
]
] |
2508.20780
|
Piotr Dybczy\'nski
|
Piotr A. Dybczy\'nski and Ma{\l}gorzata Kr\'olikowska
|
Update of the CODE catalogue and some aspects of the dynamical status of Oort Cloud comets
|
8 pages, accepted for A&A
| null | null | null |
astro-ph.EP astro-ph.IM
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Context. The outer Solar System is believed to host a vast reservoir of long-period comets (LPCs), but our understanding of their spatial distribution and dynamical history remains limited due to observational biases and uncertainties in orbital solutions for really observed comets. Aims. We aim to provide a comprehensive and dynamically homogeneous orbital database of LPCs to support the study of their origin, evolution, dynamical status, and 6D distribution of orbital elements. Methods. We updated the Catalogue of Cometary Orbits and their Dynamical Evolution (CODE catalogue) by computing original and future barycentric orbits, orbital parameters at previous and next perihelion, using full Monte Carlo swarms of real comets for the uncertainties estimation and taking into account the planetary, Galactic and passing stars perturbations according to the latest data and algorithms. Results. This update of the CODE catalogue focuses on the dynamical status of near-parabolic comets. Using current stellar data, we formulate new constraints for dynamically new comets. Now, the CODE database includes 983 orbital solutions for 369 comets with full uncertainty estimates and dynamical classifications, covering nearly all comets with original semi-major axes exceeding 10,000 au and discovered before 2022, as well as all LPCs discovered beyond 10 au from the Sun during during this period, and over 80% of the known LPCs with perihelion distances beyond 7 au.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:37:34 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Dybczyński",
"Piotr A.",
""
],
[
"Królikowska",
"Małgorzata",
""
]
] |
2508.20781
|
Jean-Guy Caputo
|
Moysey Brio and Jean-Guy Caputo
|
Numerical analysis of scattering on combinatorial graphs
| null | null | null | null |
math-ph math.MP math.SP
|
http://creativecommons.org/licenses/by/4.0/
|
We investigate numerically the scattering of waves on discrete graphs. An efficient algorithm is developed to compute the reflection and transmission (spectral) coefficients. We then explore various configurations of input and output leads, demonstrating how bound states arising from specific vertex-lead connections result in total reflection. The impedance of the leads is shown to influence the spectral coefficients in a predictable manner. Furthermore, for a given input lead we show that the total transmission of a wave packet can be maximized by appropriately selecting the exit lead. Finally, we analyze the spectral signatures of defects within the graph and find that they vary depending on both the defect's location and its spectral characteristics thus enabling its identification.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:38:15 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Brio",
"Moysey",
""
],
[
"Caputo",
"Jean-Guy",
""
]
] |
2508.20782
|
Fengyun Zhang
|
Fengyun Zhang
|
A Solution of Ultra Wideband Based High-resolution and Lossless Audio Transmission
|
5 pages
| null | null | null |
eess.AS
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
This paper provides an overview of the current challenges in wireless audio transmission and highlights the limitations of existing technologies regarding data bandwidth, data compression, latency, and inter-device compatibility. To address these shortcomings, it proposes a high-resolution, lossless audio transmission scheme utilizing ultra wideband (UWB) technology. UWB emerges as a promising solution by offering the necessary bandwidth to enable exceptional sound quality with ultra-low latency, making it ideal for real-time audio applications and addressing synchronization concerns in audio-visual use cases. Additionally, UWB's unique capabilities extend beyond high-resolution audio, allowing for precise location tracking in augmented and virtual reality applications.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:42:50 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Zhang",
"Fengyun",
""
]
] |
2508.20783
|
Beth Pearson
|
Beth Pearson, Bilal Boulbarss, Michael Wray, Martha Lewis
|
Evaluating Compositional Generalisation in VLMs and Diffusion Models
|
11 pages including references, 6 figures. Accepted at IWCS 2025
| null | null | null |
cs.CV cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
A fundamental aspect of the semantics of natural language is that novel meanings can be formed from the composition of previously known parts. Vision-language models (VLMs) have made significant progress in recent years, however, there is evidence that they are unable to perform this kind of composition. For example, given an image of a red cube and a blue cylinder, a VLM such as CLIP is likely to incorrectly label the image as a red cylinder or a blue cube, indicating it represents the image as a `bag-of-words' and fails to capture compositional semantics. Diffusion models have recently gained significant attention for their impressive generative abilities, and zero-shot classifiers based on diffusion models have been shown to perform competitively with CLIP in certain compositional tasks. In this work we explore whether the generative Diffusion Classifier has improved compositional generalisation abilities compared to discriminative models. We assess three models -- Diffusion Classifier, CLIP, and ViLT -- on their ability to bind objects with attributes and relations in both zero-shot learning (ZSL) and generalised zero-shot learning (GZSL) settings. Our results show that the Diffusion Classifier and ViLT perform well at concept binding tasks, but that all models struggle significantly with the relational GZSL task, underscoring the broader challenges VLMs face with relational reasoning. Analysis of CLIP embeddings suggests that the difficulty may stem from overly similar representations of relational concepts such as left and right. Code and dataset are available at: https://github.com/otmive/diffusion_classifier_clip
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:45:04 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Pearson",
"Beth",
""
],
[
"Boulbarss",
"Bilal",
""
],
[
"Wray",
"Michael",
""
],
[
"Lewis",
"Martha",
""
]
] |
2508.20784
|
Yifan Zhang
|
Yifan Zhang
|
Single Agent Robust Deep Reinforcement Learning for Bus Fleet Control
| null | null | null | null |
cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Bus bunching remains a challenge for urban transit due to stochastic traffic and passenger demand. Traditional solutions rely on multi-agent reinforcement learning (MARL) in loop-line settings, which overlook realistic operations characterized by heterogeneous routes, timetables, fluctuating demand, and varying fleet sizes. We propose a novel single-agent reinforcement learning (RL) framework for bus holding control that avoids the data imbalance and convergence issues of MARL under near-realistic simulation. A bidirectional timetabled network with dynamic passenger demand is constructed. The key innovation is reformulating the multi-agent problem into a single-agent one by augmenting the state space with categorical identifiers (vehicle ID, station ID, time period) in addition to numerical features (headway, occupancy, velocity). This high-dimensional encoding enables single-agent policies to capture inter-agent dependencies, analogous to projecting non-separable inputs into a higher-dimensional space. We further design a structured reward function aligned with operational goals: instead of exponential penalties on headway deviations, a ridge-shaped reward balances uniform headways and schedule adherence. Experiments show that our modified soft actor-critic (SAC) achieves more stable and superior performance than benchmarks, including MADDPG (e.g., -430k vs. -530k under stochastic conditions). These results demonstrate that single-agent deep RL, when enhanced with categorical structuring and schedule-aware rewards, can effectively manage bus holding in non-loop, real-world contexts. This paradigm offers a robust, scalable alternative to MARL frameworks, particularly where agent-specific experiences are imbalanced.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:47:40 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Zhang",
"Yifan",
""
]
] |
2508.20785
|
Abhishek Dhawan
|
Abhishek Dhawan, Eren C. K{\i}z{\i}lda\u{g}, Neeladri Maitra
|
Sharp Online Hardness for Large Balanced Independent Sets
|
31 pages; abstract shortened due to arxiv restrictions
| null | null | null |
cs.DS cs.CC cs.DM math.CO math.PR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study the algorithmic problem of finding large $\gamma$-balanced independent sets in dense random bipartite graphs; an independent set is $\gamma$-balanced if a $\gamma$ proportion of its vertices lie on one side of the bipartition. In the sparse regime, Perkins and Wang established tight bounds within the low-degree polynomial (LDP) framework, showing a factor-$1/(1-\gamma)$ statistical-computational gap via the Overlap Gap Property (OGP) framework tailored for stable algorithms. However, these techniques do not appear to extend to the dense setting. For the related large independent set problem in dense random graph, the best known algorithm is an online greedy procedure that is inherently unstable, and LDP algorithms are conjectured to fail even in the "easy" regime where greedy succeeds. We show that the largest $\gamma$-balanced independent set in dense random bipartite graphs has size $\alpha:=\frac{\log_b n}{\gamma(1-\gamma)}$ whp, where $n$ is the size of each bipartition, $p$ is the edge probability, and $b=1/(1-p)$. We design an online algorithm that achieves $(1-\epsilon)(1-\gamma)\alpha$ whp for any $\epsilon>0$. We complement this with a sharp lower bound, showing that no online algorithm can achieve $(1+\epsilon)(1-\gamma)\alpha$ with nonnegligible probability. Our results suggest that the same factor-$1/(1-\gamma)$ gap is also present in the dense setting, supporting its conjectured universality. While the classical greedy procedure on $G(n,p)$ is straightforward, our algorithm is more intricate: it proceeds in two stages, incorporating a stopping time and suitable truncation to ensure that $\gamma$-balancedness-a global constraint-is met despite operating with limited information. Our lower bound utilizes the OGP framework; we build on a recent refinement of this framework for online models and extend it to the bipartite setting.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:47:54 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Dhawan",
"Abhishek",
""
],
[
"Kızıldağ",
"Eren C.",
""
],
[
"Maitra",
"Neeladri",
""
]
] |
2508.20786
|
Kyle Ormsby
|
Caoilainn Kirkpatrick, Amelie el Mahmoud, Kyle Ormsby, Ang\'elica M. Osorno, Dale Schandelmeier-Lynch, Riley Shahar, Lixing Yi, Avery Young, Saron Zhu
|
Enumerating submonoids of finite commutative monoids
|
26 pages, 3 figures, comments welcome!
| null | null | null |
math.CO math.AT
|
http://creativecommons.org/licenses/by/4.0/
|
Given a finite commutative monoid $M$, we show that submonoids of $M\times [n]$ - where $[n] = \{0,1,\ldots,n\}$ is equipped with the max operation $\vee$ - may be enumerated via the transfer matrix method. When $M$ is also idempotent, we show that there are finitely many integers $\lambda$ and rational numbers $b_\lambda$ (only depending on $M$) such that the number of submonoids of $M\times [n]$ is $\sum_\lambda b_\lambda\lambda^n$. This answers a question of Knuth regarding ternary (and higher order) max-closed relations, and has applications to the enumeration of saturated transfer systems in equivariant infinite loop space theory.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:48:07 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Kirkpatrick",
"Caoilainn",
""
],
[
"Mahmoud",
"Amelie el",
""
],
[
"Ormsby",
"Kyle",
""
],
[
"Osorno",
"Angélica M.",
""
],
[
"Schandelmeier-Lynch",
"Dale",
""
],
[
"Shahar",
"Riley",
""
],
[
"Yi",
"Lixing",
""
],
[
"Young",
"Avery",
""
],
[
"Zhu",
"Saron",
""
]
] |
2508.20787
|
Reinhold Willcox
|
R. Willcox, F. R. N. Schneider, E. Laplace, Ph. Podsiadlowski, K. Maltsev, I. Mandel, P. Marchant, H. Sana, T. Li, T. Hertog
|
New gravitational-wave data support a bimodal black-hole mass distribution
| null | null | null | null |
astro-ph.HE gr-qc
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Detailed stellar evolution and supernova models yield a bimodal black-hole mass distribution with a narrow peak around 10 solar masses from stars within a narrow range of progenitor properties and a second broader peak starting around 20 solar masses from very massive progenitors. This bimodal black-hole mass distribution leads to a characteristic distribution of chirp masses of merging binary black holes, with two main peaks arising from the merger of two black holes where both come either from the low- or the high-mass peak and a smaller peak in between from the mixed merger of a low-mass and a high-mass black hole. We carry out a population synthesis study of binary black hole formation and compare the results to the observed chirp masses of gravitational-wave events. We find that only the bimodal black-hole mass prescription is able to reproduce the structure of peaks and gaps in the observed chirp-mass distribution, which is not matched by predictions from other remnant mass prescriptions in the literature.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:52:12 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Willcox",
"R.",
""
],
[
"Schneider",
"F. R. N.",
""
],
[
"Laplace",
"E.",
""
],
[
"Podsiadlowski",
"Ph.",
""
],
[
"Maltsev",
"K.",
""
],
[
"Mandel",
"I.",
""
],
[
"Marchant",
"P.",
""
],
[
"Sana",
"H.",
""
],
[
"Li",
"T.",
""
],
[
"Hertog",
"T.",
""
]
] |
2508.20788
|
Henryk Gzyl
|
Henryk Gzyl
|
A predictive solution of the EPR paradox
| null | null | null | null |
quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
In this work an incorrect argument in EPR's paper is corrected. A predictive approach to further confirm the validity of quantum theory is also proposed. The essence of the detail that EPR missed is that in a state of given total momentum (in their example the total momentum is zero), since the total momentum operator $\hat{\bp}=\hat{\bp}_1+\hat{\bp}_2$ does not commute with any of the position operators $\hat{\bx}_1$ and $\hat{\bx}_2,$ then in an eigenstate of the total momentum operator, the standard deviation on the measurement of the position of any of the two particles has to be infinite. Below we compute the (quantum) conditional expectation of the momentum of any of the particles (say, the first) given the value of a measurement of the total momentum. Since the predictor of the momentum of the particle depends on the total momentum, and the total momentum does not commute with the position operator of any of the particles, the uncertainty principle is not violated, and no contradiction appears. We also prove, using the basic tenets of quantum measurement theory, that after measurement of the total momentum, the probability density (computed as the absolute square of the wave function), coincides with the (quantum) conditional density given the total momentum.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:53:03 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Gzyl",
"Henryk",
""
]
] |
2508.20789
|
Xueyang Kang Mr.
|
Xueyang Kang, Hang Zhao, Kourosh Khoshelham and Patrick Vandewalle
|
Surfel-based 3D Registration with Equivariant SE(3) Features
|
5 pages, 4 figures
|
Published on 2025 IEEE International Geoscience and Remote Sensing Symposium
| null | null |
cs.CV cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Point cloud registration is crucial for ensuring 3D alignment consistency of multiple local point clouds in 3D reconstruction for remote sensing or digital heritage. While various point cloud-based registration methods exist, both non-learning and learning-based, they ignore point orientations and point uncertainties, making the model susceptible to noisy input and aggressive rotations of the input point cloud like orthogonal transformation; thus, it necessitates extensive training point clouds with transformation augmentations. To address these issues, we propose a novel surfel-based pose learning regression approach. Our method can initialize surfels from Lidar point cloud using virtual perspective camera parameters, and learns explicit $\mathbf{SE(3)}$ equivariant features, including both position and rotation through $\mathbf{SE(3)}$ equivariant convolutional kernels to predict relative transformation between source and target scans. The model comprises an equivariant convolutional encoder, a cross-attention mechanism for similarity computation, a fully-connected decoder, and a non-linear Huber loss. Experimental results on indoor and outdoor datasets demonstrate our model superiority and robust performance on real point-cloud scans compared to state-of-the-art methods.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:53:44 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Kang",
"Xueyang",
""
],
[
"Zhao",
"Hang",
""
],
[
"Khoshelham",
"Kourosh",
""
],
[
"Vandewalle",
"Patrick",
""
]
] |
2508.20790
|
Peter Woitke
|
Peter Woitke, Manuel Scherf, Christiane Helling, Paul B. Rimmer, Martin Ferus, Helmut Lammer, Fabian Weichbold, Kate\v{r}ina N\v{e}me\v{c}kov\'a, Petr Eminger, Jaroslav Ka\v{c}ina, Tereza Constantinou
|
Prediction of sulphate hazes in the lower Venus atmosphere
|
18 pages, accepted by PSJ on 27-Aug-2025
| null | null | null |
astro-ph.EP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study the amount, size distribution and material composition of (sub-)mic aerosol particles in the lower Venus atmosphere < 50 km. Our GGchem phase-equilibrium model predicts metal-chloride and metal-fluoride molecules to be present in the gas over the Venus surface in trace concentrations < 2.E-12, in particular FeCl2, NaCl, KCl and SiF4. Using an improved version of the DiffuDrift model developed by Woitke et al.2020, we find that these molecules deposit to form solid potassium sulphate K2SO4, sodium sulphate Na2SO4, and pyrite FeS2 above about 15.5 km, 9.5 km and 2.4 km, respectively. These heights coincide well with the three potential haze layers found in the Pioneer Venus Large Probe neutral mass spectrometer data by Mogul et al.2023. The particles with radius < 0.3 mic can be dredged up from the ground to reach the sulphuric acid cloud base from below by diffusion. The particle density decreases from ~ 5000/cm3 at ground level to ~100/cm3 at a height of 45 km. Particles larger than about 1 mic are found to stay confined to the ground < 10 km, indicating that the larger, so-called mode 3 particles, if they exist, cannot originate from the surface. All particles are expected to be coated by a thin layer of FeS2, Na2SO4 and K2SO4. We have included the repelling effect of particle charges on the coagulation, without which the model would predict much too steep gradients close to the surface, which is inconsistent with measured opacity data. Our models suggest that the particles must have at least 100 negative charges per micron of particle radius at ground level, and > 50/mic at a height of 45 km.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:54:19 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Woitke",
"Peter",
""
],
[
"Scherf",
"Manuel",
""
],
[
"Helling",
"Christiane",
""
],
[
"Rimmer",
"Paul B.",
""
],
[
"Ferus",
"Martin",
""
],
[
"Lammer",
"Helmut",
""
],
[
"Weichbold",
"Fabian",
""
],
[
"Němečková",
"Kateřina",
""
],
[
"Eminger",
"Petr",
""
],
[
"Kačina",
"Jaroslav",
""
],
[
"Constantinou",
"Tereza",
""
]
] |
2508.20791
|
Katarzyna Mazurek
|
P. Jucha, K. Mazurek, A. Szczurek, and K. Pysz
|
New UPC results at the LHC - modeling the proton and neutron emission in photon-induced nuclear processes
|
20 pages, 12 figures
| null | null | null |
hep-ph nucl-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The ultrarelativistic collisions of heavy ions provide rich spectrum of possibilities to discuss the response of the nucleus to photons. Newly published neutron and proton multiplicities measured in the ALICE experiment in ultraperipheral collisions allow investigating the influence of the electromagnetic fields on colliding nuclei for the $^{208}$Pb+$^{208}$Pb at $\sqrt{s_{NN}}$=5.02~TeV. The theoretical predictions are done within hybrid model including equivalent photon approximations (EPA), GiBUU modeling of pre-equilibrium processes and generation of the exited nuclear remnants, which decay is modeled by statistical approach: GEM2 or GEMINI++. The cross-sections of the mass-charge distributions of nuclear remnants as well as the neutron, proton and other charged particle multiplicities are estimated. We concentrate on production of protons and isotopes coming from the electromagnetic dissociation. Special attention is devoted to emission of a single proton. The cross section for $1p$ emission is very close to maximal available one based on reactions of photon with individual nucleons. Our pre-equilibrium processes explain simultaneously the tail of neutron energy distributions in the nuclear rest frame observed in $\gamma + A$ collisions.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:54:22 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Jucha",
"P.",
""
],
[
"Mazurek",
"K.",
""
],
[
"Szczurek",
"A.",
""
],
[
"Pysz",
"K.",
""
]
] |
2508.20792
|
Rajae Malek
|
Rajae Malek, Qing-Dong Jiang, Haiwen Liu
|
Distinct Spatiotemporal Dynamics of Thermoelectric Transport Across Superconducting Transition
| null | null | null | null |
cond-mat.supr-con
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We investigate the relaxation dynamics of heat transport in superconductors, shaped by the interplay of diffusion, nonlinearity, and magnetic fields. Focusing on regimes near the critical temperature Tc, we analyze two classes of relaxation diffusion equations that give rise to qualitatively distinct dynamics, which we denote as Type I (linear) and Type II (nonlinear). Type I relaxation, characteristic of the normal state above Tc, results in a steady and spatially uniform heat current governed by linear diffusion. By contrast, Type II relaxation, relevant below Tc, exhibits non steady dynamics marked by pronounced spatial inhomogeneities and an evolving pattern, in which an initially localized hot spot propagates transiently through the system. The striking distinction between these regimes underscores a fundamental shift in transport mechanisms across the superconducting phase transition and provides experimentally relevant predictions in light of emerging techniques for probing local dissipation.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:54:36 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Malek",
"Rajae",
""
],
[
"Jiang",
"Qing-Dong",
""
],
[
"Liu",
"Haiwen",
""
]
] |
2508.20793
|
Tom\'a\v{s} Proch\'azka
|
Mat\v{e}j Kudrna, Tom\'a\v{s} Proch\'azka
|
On W-algebras and ODE/IM correspondence
| null | null | null | null |
hep-th math-ph math.MP math.QA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study the ODE/IM correspondence for two-dimensional conformal field theories with Virasoro and $\mathcal{W}_N$ symmetry. Building on earlier work establishing the correspondence, we develop a systematic algorithm for calculating the eigenvalues of local integrals of motion in terms of the Bethe roots using formal WKB expansions of wave functions associated to the differential operators. The method is demonstrated explicitly for Virasoro, $\mathcal{W}_3$, and $\mathcal{W}_4$ algebras, yielding closed expressions for the eigenvalues of the first few local quantum KdV Hamiltonians. A key geometric structure emerging from our analysis is the mirror curve, a three-punctured sphere that is naturally covered by the WKB curve. We show how the algebraic properties of the $\mathcal{W}$-symmetry algebras are reflected in the geometry of these curves, and how period integrals on these curves reproduce the spectral data of the integrable system. Applications to Argyres-Douglas minimal models allow us to test the prescription both analytically and numerically and we find complete agreement between the calculations in different triality frames. Finally, we examine large rank limits of ground state eigenvalues and show that they match the genus expansion of the topological string partition function on $\mathbb{C}^3$.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:54:47 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Kudrna",
"Matěj",
""
],
[
"Procházka",
"Tomáš",
""
]
] |
2508.20794
|
Sofie Ried
|
Sofie Ried
|
Singularity Resolution of Quantum Black Holes in (A)dS
|
4 pages, 1 figure
| null | null | null |
gr-qc hep-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The singularities present at the centre of black holes signal a break down of the classical theory. In this paper, we demonstrate a resolution of the Schwarzschild-(Anti-)de Sitter singularity by imposing unitary evolution with respect to unimodular time. Employing the Henneaux-Teitelboim formulation of unimodular gravity, we perform a canonical quantization on a symmetry-reduced Schwarzschild-(Anti-) de Sitter model. This leads to a Wheeler-DeWitt equation that effectively becomes a Schr\"odinger equation in unimodular time. By imposing unitarity, we discover a family of quantum theories in which the classical singularity is resolved. These theories each allow only semiclassical states corresponding to one mass sign: either positive, negative, or zero. Furthermore, we derive an analytical expression for the quantum-corrected Schwarzschild metric, which is modified by a new length scale $r_{min}$ that governs the black hole's transition to a white hole.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:57:01 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Ried",
"Sofie",
""
]
] |
2508.20795
|
Marcelo Medeiros
|
Marcelo C. Medeiros and Jeronymo M. Pinro
|
Time Series Embedding and Combination of Forecasts: A Reinforcement Learning Approach
| null | null | null | null |
econ.EM
|
http://creativecommons.org/licenses/by/4.0/
|
The forecasting combination puzzle is a well-known phenomenon in forecasting literature, stressing the challenge of outperforming the simple average when aggregating forecasts from diverse methods. This study proposes a Reinforcement Learning - based framework as a dynamic model selection approach to address this puzzle. Our framework is evaluated through extensive forecasting exercises using simulated and real data. Specifically, we analyze the M4 Competition dataset and the Survey of Professional Forecasters (SPF). This research introduces an adaptable methodology for selecting and combining forecasts under uncertainty, offering a promising advancement in resolving the forecasting combination puzzle.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:58:09 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Medeiros",
"Marcelo C.",
""
],
[
"Pinro",
"Jeronymo M.",
""
]
] |
2508.20796
|
Xiaoxiao Miao
|
ChenYi Chua, JunKai Wong, Chengxin Chen, Xiaoxiao Miao
|
Speech Emotion Recognition via Entropy-Aware Score Selection
|
The paper has been accepted by APCIPA ASC 2025
| null | null | null |
cs.SD cs.AI
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
In this paper, we propose a multimodal framework for speech emotion recognition that leverages entropy-aware score selection to combine speech and textual predictions. The proposed method integrates a primary pipeline that consists of an acoustic model based on wav2vec2.0 and a secondary pipeline that consists of a sentiment analysis model using RoBERTa-XLM, with transcriptions generated via Whisper-large-v3. We propose a late score fusion approach based on entropy and varentropy thresholds to overcome the confidence constraints of primary pipeline predictions. A sentiment mapping strategy translates three sentiment categories into four target emotion classes, enabling coherent integration of multimodal predictions. The results on the IEMOCAP and MSP-IMPROV datasets show that the proposed method offers a practical and reliable enhancement over traditional single-modality systems.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:58:09 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Chua",
"ChenYi",
""
],
[
"Wong",
"JunKai",
""
],
[
"Chen",
"Chengxin",
""
],
[
"Miao",
"Xiaoxiao",
""
]
] |
2508.20797
|
Fei Wei Dr.
|
Peter J. Forrester and Fei Wei
|
Higher order linear differential equations for unitary matrix integrals: applications and generalisations
|
18 pages
| null | null | null |
math-ph math.MP
|
http://creativecommons.org/licenses/by/4.0/
|
We consider characterisations of the class of unitary matrix integrals $ \langle (\det U)^q e^{s^{1/2} {\rm Tr}(U + U^\dagger)} \rangle_{U(l)}$ in terms of a first order matrix linear differential equation for a vector function of size $l+1$, and in terms of a scalar linear differential equation of degree $l+1$. It will be shown that the latter follows from the former. The matrix linear differential equation provides an efficient way to compute the power series expansion of the matrix integrals, which with $q=0$ and $q=l$ are of relevance to the enumeration of longest increasing subsequences for random permutations, and to the question of the moments of the first and second derivative of the Riemann zeta on the critical line respectively. This procedure is compared against that following from known characterisations involving the $\sigma$-Painlev\'e III$'$ second order nonlinear differential equation. We show too that that the natural $\beta$ generalisation of the unitary group integral permits characterisation by the same classes of linear differential equations.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:58:15 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Forrester",
"Peter J.",
""
],
[
"Wei",
"Fei",
""
]
] |
2508.20798
|
Zechun Niu
|
Zechun Niu, Lang Mei, Liu Yang, Ziyuan Zhao, Qiang Yan, Jiaxin Mao, and Ji-Rong Wen
|
Addressing Personalized Bias for Unbiased Learning to Rank
|
Accepted by CIKM 2025
| null |
10.1145/3746252.3761377
| null |
cs.IR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Unbiased learning to rank (ULTR), which aims to learn unbiased ranking models from biased user behavior logs, plays an important role in Web search. Previous research on ULTR has studied a variety of biases in users' clicks, such as position bias, presentation bias, and outlier bias. However, existing work often assumes that the behavior logs are collected from an ``average'' user, neglecting the differences between different users in their search and browsing behaviors. In this paper, we introduce personalized factors into the ULTR framework, which we term the user-aware ULTR problem. Through a formal causal analysis of this problem, we demonstrate that existing user-oblivious methods are biased when different users have different preferences over queries and personalized propensities of examining documents. To address such a personalized bias, we propose a novel user-aware inverse-propensity-score estimator for learning-to-rank objectives. Specifically, our approach models the distribution of user browsing behaviors for each query and aggregates user-weighted examination probabilities to determine propensities. We theoretically prove that the user-aware estimator is unbiased under some mild assumptions and shows lower variance compared to the straightforward way of calculating a user-dependent propensity for each impression. Finally, we empirically verify the effectiveness of our user-aware estimator by conducting extensive experiments on two semi-synthetic datasets and a real-world dataset.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:01:31 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Niu",
"Zechun",
""
],
[
"Mei",
"Lang",
""
],
[
"Yang",
"Liu",
""
],
[
"Zhao",
"Ziyuan",
""
],
[
"Yan",
"Qiang",
""
],
[
"Mao",
"Jiaxin",
""
],
[
"Wen",
"Ji-Rong",
""
]
] |
2508.20799
|
Tom Lenaerts
|
Marco Saponara and Elias Fernandez Domingos and Jorge M. Pacheco and Tom Lenaerts
|
Evolution favours positively biased reasoning in sequential interactions with high future gains
|
33 pages, 5 figures
|
Saponara Marco, Fernandez Domingos Elias, Pacheco Jorge M. and Lenaerts Tom 2025 Evolution favours positively biased reasoning in sequential interactions with high future gains J. R. Soc. Interface.2220250153
|
10.1098/rsif.2025.0153
| null |
cs.MA
|
http://creativecommons.org/licenses/by/4.0/
|
Empirical evidence shows that human behaviour often deviates from game-theoretical rationality. For instance, humans may hold unrealistic expectations about future outcomes. As the evolutionary roots of such biases remain unclear, we investigate here how reasoning abilities and cognitive biases co-evolve using Evolutionary Game Theory. In our model, individuals in a population deploy a variety of unbiased and biased level-k reasoning strategies to anticipate others' behaviour in sequential interactions, represented by the Incremental Centipede Game. Positively biased reasoning strategies have a systematic inference bias towards higher but uncertain rewards, while negatively biased strategies reflect the opposite tendency. We find that selection consistently favours positively biased reasoning, with rational behaviour even going extinct. This bias co-evolves with bounded rationality, as the reasoning depth remains limited in the population. Interestingly, positively biased agents may co-exist with non-reasoning agents, thus pointing to a novel equilibrium. Longer games further promote positively biased reasoning, as they can lead to higher future rewards. The biased reasoning strategies proposed in this model may reflect cognitive phenomena like wishful thinking and defensive pessimism. This work therefore supports the claim that certain cognitive biases, despite deviating from rational judgment, constitute an adaptive feature to better cope with social dilemmas.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:01:54 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Saponara",
"Marco",
""
],
[
"Domingos",
"Elias Fernandez",
""
],
[
"Pacheco",
"Jorge M.",
""
],
[
"Lenaerts",
"Tom",
""
]
] |
2508.20800
|
James Maxwell
|
James D. Maxwell
|
EPICS for Small-Scale Laboratories with Python Soft IOCs
| null | null | null |
JLAB-PHY-25-4495
|
physics.ins-det
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
While the Experimental Physics and Industrial Control System (EPICS) is widely used at large laboratories for slow controls and instrumentation, the deployment of a full EPICS installation can be difficult, with a steep learning curve to new users. Taking advantage of the pythonSoftIOC module, we developed an EPICS slow controls implementation for Jefferson Lab's Hall B cryotarget written entirely in Python and based on software IOCs that communicate with instruments over Ethernet. This system ran successfully, interfacing with Jefferson Lab's full EPICS network, and we offer it as an example of the capabilities of pythonSoftIOC to build lightweight, yet robust and flexible instrumentation platforms that would be easily adapted for use at a small-scale laboratory. University groups can use these examples to build complete slow controls systems, from device communication to data archiving and display, using open-source, mature EPICS tools and student-friendly Python as an alternative to expensive and proprietary systems such as LabView.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:01:54 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Maxwell",
"James D.",
""
]
] |
2508.20801
|
Matthew Shirley
|
Matthew D. Shirley, Ian M. Griffiths, Joseph Houghton, Lucy Hope, Paul Hope, Christopher. J. W. Breward
|
Mathematical Modelling of Face Coverings for Virus Protection
| null | null | null | null |
physics.flu-dyn
|
http://creativecommons.org/licenses/by/4.0/
|
Traditional face masks used extensively, for example, during the COVID-19 pandemic, are relatively impermeable, forcing aerosol-containing breath to flow into the surrounding atmosphere by ``leaking" out from the lateral edges of the mask. This protects close-proximity persons but results in a high viral load being delivered into the air. An alternative face covering, the Virustatic Shield, developed by Virustatic, involves a much more permeable material coated in an active agent, with the goal of increasing the transmission through the mask and thus reducing the viral load released. In this paper, we build and solve a model that describes the air flow and aerosol transport through a porous face mask. Our model for air flow is based on a low-Reynolds-number approximation to the Navier--Stokes equations, and reduces to a two-dimensional scalar partial differential equation for the local flux through the mask. We use the model to explore the interplay between leakage from the mask sides and flow directly through. We compare and contrast the leakage for traditional face masks and for neck gaiters, and find that neck gaiters have a lower leakage. We utilise a simple model for aerosol transport that balances advection with capture by the mask, and find that there is an ideal permeability, which optimises real-world protection offered by the mask.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:02:14 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Shirley",
"Matthew D.",
""
],
[
"Griffiths",
"Ian M.",
""
],
[
"Houghton",
"Joseph",
""
],
[
"Hope",
"Lucy",
""
],
[
"Hope",
"Paul",
""
],
[
"Breward",
"Christopher. J. W.",
""
]
] |
2508.20802
|
Lev Yungelson
|
Mikhail Prokhorov, Kefeng Tan, Nikolay Samus, Ali Luo, Dana Kovaleva, Jingkun Zhao, Yujuan Liu, Pavel Kaygorodov, Oleg Malkov, Yihan Song, Sergey Sichevskij, Lev Yungelson, and Gang Zhao
|
Looking for Signs of Unresolved Binarity in the Continuum of LAMOST Stellar Spectra
|
29 p.p., 5 Figures
|
Galaxies 2025, 13, 83
|
10.3390/galaxies13040083
| null |
astro-ph.SR astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
We describe an attempt to derive the binarity rate of samples of 166 A-, F-, G-, and K-type stars from LAMOST DR5 and 1000 randomly selected presumably single stars from Gaia DR3 catalogs. To this end, we compared continua of the observed spectra with the continua of synthetic spectra in the 3700 to 9097 Angstrom range. The latter spectra were reduced to the LAMOST set of wavelengths, while the former ones were smoothed. Next, we searched for every observed star the nearest synthetic spectrum using a four-parameter representation - effective temperature, gravity, [Fe/H], and a range of interstellar absorption values. However, rms deviations of observed spectra from synthetic ones appeared to be not sufficient to claim that any of the stars is a binary. We conclude that comparison of the intensity of pairs of spectral lines remains the best way to detect binarity.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:05:39 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Prokhorov",
"Mikhail",
""
],
[
"Tan",
"Kefeng",
""
],
[
"Samus",
"Nikolay",
""
],
[
"Luo",
"Ali",
""
],
[
"Kovaleva",
"Dana",
""
],
[
"Zhao",
"Jingkun",
""
],
[
"Liu",
"Yujuan",
""
],
[
"Kaygorodov",
"Pavel",
""
],
[
"Malkov",
"Oleg",
""
],
[
"Song",
"Yihan",
""
],
[
"Sichevskij",
"Sergey",
""
],
[
"Yungelson",
"Lev",
""
],
[
"Zhao",
"Gang",
""
]
] |
2508.20803
|
Jiahui Zhang
|
Chunjing Li, Jiahui Zhang, Xiaohui Yuan
|
Optional subsampling for generalized estimating equations in growing-dimensional longitudinal Data
|
34 pages, 5 figures
| null | null | null |
stat.CO stat.ME
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
As a powerful tool for longitudinal data analysis, the generalized estimating equations have been widely studied in the academic community. However, in large-scale settings, this approach faces pronounced computational and storage challenges. In this paper, we propose an optimal Poisson subsampling algorithm for generalized estimating equations in large-scale longitudinal data with diverging covariate dimension, and establish the asymptotic properties of the resulting estimator. We further derive the optimal Poisson subsampling probability based on A- and L-optimality criteria. An approximate optimal Poisson subsampling algorithm is proposed, which adopts a two-step procedure to construct these probabilities. Simulation studies are conducted to evaluate the performance of the proposed method under three different working correlation matrices. The results show that the method remains effective even when the working correlation matrices are misspecified. Finally, we apply the proposed method to the CHFS dataset to illustrate its empirical performance.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:06:36 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Li",
"Chunjing",
""
],
[
"Zhang",
"Jiahui",
""
],
[
"Yuan",
"Xiaohui",
""
]
] |
2508.20804
|
Lucas Amoudruz
|
Lucas Amoudruz, Gregory Buti, Luciano Rivetti, Ali Ajdari, Gregory Sharp, Petros Koumoutsakos, Simon Spohn, Anca L Grosu, Thomas Bortfeld
|
Ising energy model for the stochastic prediction of tumor islets
| null | null | null | null |
physics.med-ph physics.comp-ph q-bio.QM
|
http://creativecommons.org/licenses/by/4.0/
|
A major challenge in diagnosing and treating cancer is the infiltrative growth of tumors into surrounding tissues.
This microscopic spread of the disease is invisible on most diagnostic imaging modalities and can often only be detected histologically in biopsies.
The purpose of this paper is to develop a physically based model of tumor spread that captures the histologically observed behavior in terms of seeding small tumor islets in prostate cancer.
The model is based on three elementary events: a tumor cell can move, duplicate, or die.
The propensity of each event is given by an Ising-like Hamiltonian that captures correlations between neighboring cells.
The model parameters were fitted to clinical data obtained from surgical specimens taken from 23 prostate cancer patients.
The results demonstrate that this straightforward physical model effectively describes the distribution of the size and the number of tumor islets in prostate cancer.
The simulated tumor islets exhibit a regular, approximately spherical shape, correctly mimicking the shapes observed in histology.
This is due to the Ising interaction term between neighboring cells acting as a surface tension that gives rise to regularly shaped islets.
The model addresses the important clinical need of calculating the probability of tumor involvement in specific sub-volumes of the prostate, which is required for radiation treatment planning and other applications.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:06:38 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Amoudruz",
"Lucas",
""
],
[
"Buti",
"Gregory",
""
],
[
"Rivetti",
"Luciano",
""
],
[
"Ajdari",
"Ali",
""
],
[
"Sharp",
"Gregory",
""
],
[
"Koumoutsakos",
"Petros",
""
],
[
"Spohn",
"Simon",
""
],
[
"Grosu",
"Anca L",
""
],
[
"Bortfeld",
"Thomas",
""
]
] |
2508.20805
|
Xiaoxiao Miao
|
Javier Si Zhao Hong, Timothy Zoe Delaya, Sherwyn Chan Yin Kit, Pai Chet Ng, Xiaoxiao Miao
|
Exploring Machine Learning and Language Models for Multimodal Depression Detection
|
This paper has been accepted by APCIPA ASC 2025
| null | null | null |
cs.CL cs.AI cs.SD
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
This paper presents our approach to the first Multimodal Personality-Aware Depression Detection Challenge, focusing on multimodal depression detection using machine learning and deep learning models. We explore and compare the performance of XGBoost, transformer-based architectures, and large language models (LLMs) on audio, video, and text features. Our results highlight the strengths and limitations of each type of model in capturing depression-related signals across modalities, offering insights into effective multimodal representation strategies for mental health prediction.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:07:07 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Hong",
"Javier Si Zhao",
""
],
[
"Delaya",
"Timothy Zoe",
""
],
[
"Kit",
"Sherwyn Chan Yin",
""
],
[
"Ng",
"Pai Chet",
""
],
[
"Miao",
"Xiaoxiao",
""
]
] |
2508.20806
|
Moriba Jah
|
Moriba Jah, Van Haslett
|
The Epistemic Support-Point Filter (ESPF): A Bounded Possibilistic Framework for Ordinal State Estimation
| null | null | null | null |
cs.IT cs.CE cs.SY eess.SY math.DS math.IT
|
http://creativecommons.org/licenses/by/4.0/
|
Traditional state estimation methods rely on probabilistic assumptions that often collapse epistemic uncertainty into scalar beliefs, risking overconfidence in sparse or adversarial sensing environments. We introduce the Epistemic Support-Point Filter (ESPF), a novel non-Bayesian filtering framework fully grounded in possibility theory and epistemic humility. ESPF redefines the evolution of belief over state space using compatibility-weighted support updates, surprisalaware pruning, and adaptive dispersion via sparse grid quadrature. Unlike conventional filters, ESPF does not seek a posterior distribution, but rather maintains a structured region of plausibility or non-rejection, updated using ordinal logic rather than integration. For multi-model inference, we employ the Choquet integral to fuse competing hypotheses based on a dynamic epistemic capacity function, generalizing classical winner-take-all strategies. The result is an inference engine capable of dynamically contracting or expanding belief support in direct response to information structure, without requiring prior statistical calibration. This work presents a foundational shift in how inference, evidence, and ignorance are reconciled, supporting robust estimation where priors are unavailable, misleading, or epistemically unjustified.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:08:13 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Jah",
"Moriba",
""
],
[
"Haslett",
"Van",
""
]
] |
2508.20807
|
Ke Xu
|
Ke Xu, Fen-Fen Yang, Chenggui Yuan
|
On the infinite time horizon approximation for L\'evy-driven McKean-Vlasov SDEs with common noise
| null | null | null | null |
math.PR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this work, we establish the existence and uniqueness of solutions to McKean-Vlasov stochastic differential equations (SDEs) driven by L\'evy processes with common noise on an infinite time horizon, by means of a contraction mapping principle in the space of probability measures. In addition, we analyse the propagation of chaos for L\'evy-driven McKean-Vlasov SDEs in the presence of common noise.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:08:46 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Xu",
"Ke",
""
],
[
"Yang",
"Fen-Fen",
""
],
[
"Yuan",
"Chenggui",
""
]
] |
2508.20808
|
Alice Publications
|
ALICE Collaboration
|
$\Sigma^{+}$ production in pp collisions at $\sqrt{s} = 13$ TeV
|
22 pages, 5 captioned figures, 3 tables, authors from page 16, submitted to EPJC, figures at http://alice-publications.web.cern.ch/node/12756
| null | null |
CERN-EP-2025-181
|
nucl-ex hep-ex
|
http://creativecommons.org/licenses/by/4.0/
|
The measurement of $\Sigma^{+}$ production in pp collisions at $\sqrt{s} = 13$ TeV is presented. The measurement is performed at midrapidity in both minimum-bias and high-multiplicity pp collisions at $\sqrt{s} = 13$ TeV. The $\Sigma^{+}$ is reconstructed via its weak-decay topology in the decay channel $\Sigma^{+} \rightarrow {\rm p} + \pi^{0}$ with $\pi^{0} \rightarrow \gamma + \gamma$. In a novel approach, the neutral pion is reconstructed by combining photons that convert in the detector material with photons measured in the calorimeters. The transverse-momentum ($p_{\rm T}$) distributions of the $\Sigma^{+}$ and its rapidity densities ${\rm d}N$/${\rm d}y$ in both event classes are reported. The $p_{\rm T}$ spectrum in minimum-bias collisions is compared to QCD-inspired event generators. The ratio of $\Sigma^{+}$ to previously measured $\Lambda$ baryons is in good agreement with calculations from the Statistical Hadronization Model. The high efficiency and purity of the novel reconstruction method for $\Sigma^{+}$ presented here will enable future studies of the interaction of $\Sigma^{+}$ with protons in the context of femtoscopic measurements, which could be crucial for understanding the equation of state of neutron stars.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:09:01 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"ALICE Collaboration",
"",
""
]
] |
2508.20809
|
Wen-Hui Ai
|
Jia-Long Chen and Wen-Hui Ai
|
On the Spectral Properties of a Class of Planar Sierpinski Self-Affine Measures
|
32pages
| null | null | null |
math.FA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We investigate the spectral properties of a class of Sierpinski-type self-affine measures defined by
\[
\mu_{M,D}(\cdot) = p^{-1} \sum_{d \in D} \mu_{M,D}(M(\cdot) - d),
\]
where \( p \) is a prime number, \( M = \begin{bmatrix}
\rho_1^{-1} & c
0 & \rho_2^{-1}
\end{bmatrix} \) is a real upper triangular expanding matrix, and \( D = \{d_0, d_1, \cdots, d_{p-1}\} \subset \mathbb{Z}^2 \) satisfying \( \mathcal{Z}(\widehat{\delta}_{D}) = \cup_{j=1}^{p-1} \left( \frac{j \bm{a}}{p} + \mathbb{Z}^2 \right) \) for some \( \bm{a} \in \mathcal{E}_{p}= \{ (i_1, i_2)^* : i_1, i_2 \in [1, p-1] \cap \mathbb{Z} \} \), where \( \mathcal{Z}(\widehat{\delta}_{D}) \) denotes the set of zeros of \( \widehat{\delta}_{D} \) with \( \delta_{D} = \frac{1}{\# D} \sum_{d \in D} \delta_d \). When $\rho_1 = \rho_2$, we derive necessary and sufficient conditions for $\mu_{M,D}$ to both: $(i)$ possess an infinite orthogonal set of exponential functions, and $(ii)$ be a spectral measure. When no infinite orthogonal exponential system exists in $L^{2}(\mu_{M,D})$, we quantify the maximum number of orthogonal exponentials and provide precise estimates. For $\rho_1 \neq \rho_2$, with restricted digit sets $D$, we obtain a necessary and sufficient condition for $\mu_{M,D}$ to be a spectral measure.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:10:46 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Chen",
"Jia-Long",
""
],
[
"Ai",
"Wen-Hui",
""
]
] |
2508.20810
|
Jessica Lundin PhD
|
Jessica Lundin, Guillaume Chabot-Couture
|
A Graph-Based Test-Harness for LLM Evaluation
|
4 pages, 2 figures, dataset
| null | null | null |
cs.AI cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
We present a first known prototype of a dynamic, systematic benchmark of medical guidelines for 400+ questions, with 3.3+ trillion possible combinations, covering 100\% of guideline relationships. We transformed the WHO IMCI handbook into a directed graph with 200+ nodes (conditions, symptoms, treatments, follow-ups, severities) and 300+ edges, then used graph traversal to generate questions that incorporated age-specific scenarios and contextual distractors to ensure clinical relevance. Our graph-based approach enables systematic evaluation across clinical tasks (45-67\% accuracy), and we find models excel at symptom recognition but struggle with triaging severity, treatment protocols and follow-up care, demonstrating how customized benchmarks can identify specific capability gaps that general-domain evaluations miss. Beyond evaluation, this dynamic MCQA methodology enhances LLM post-training (supervised finetuning, GRPO, DPO), where correct answers provide high-reward samples without expensive human annotation. The graph-based approach successfully addresses the coverage limitations of manually curated benchmarks. This methodology is a step toward scalable, contamination-resistant solution for creating comprehensive benchmarks that can be dynamically generated, including when the guidelines are updated. Code and datasets are available at https://github.com/jessicalundin/graph_testing_harness
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:10:59 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Lundin",
"Jessica",
""
],
[
"Chabot-Couture",
"Guillaume",
""
]
] |
2508.20811
|
Tita Alissa Bach
|
Tita Alissa Bach, Linn Pedersen, Maria Kinck Bor\'en{\dag}, Lisa Christoffersen Temte{\dag}
|
When technology is not enough: Insights from a pilot cybersecurity culture assessment in a safety-critical industrial organisation
|
*Group Research and Development, DNV, 1363 H{\o}vik, Norway, {\dag}Safety Risk Nordic, DNV, 1363 H{\o}vik, Norway
| null | null | null |
cs.CY
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
As cyber threats increasingly exploit human behaviour, technical controls alone cannot ensure organisational cybersecurity (CS). Strengthening cybersecurity culture (CSC) is vital in safety-critical industries, yet empirical research in real-world industrial setttings is scarce. This paper addresses this gap through a pilot mixed-methods CSC assessment in a global safety-critical organisation. We examined employees' CS knowledge, attitudes, behaviours, and organisational factors shaping them. A survey and semi-structured interviews were conducted at a global organisation in safety-critical industries, across two countries chosen for contrasting phishing simulation performance: Country 1 stronger, Country 2 weaker. In Country 1, 258 employees were invited (67%), in Country 2, 113 were invited (30%). Interviews included 20 and 10 participants respectively. Overall CSC profiles were similar but revealed distinct challenges. Both showed strong phishing awareness and prioritised CS, yet most viewed phishing as the main risk and lacked clarity on handling other incidents. Line managers were default contacts, but follow-up on reported concerns was unclear. Participants emphasized aligning CS expectations with job relevance and workflows. Key contributors to differences emerged: Country 1 had external employees with limited access to CS training and policies, highlighting monitoring gaps. In Country 2, low survey response stemmed from a "no-link in email" policy. While this policy may have boosted phishing performance, it also underscored inconsistencies in CS practices. Findings show that resilient CSC requires leadership involvement, targeted communication, tailored measures, policy-practice alignment, and regular assessments. Embedding these into strategy complements technical defences and strengthens sustainable CS in safety-critical settings.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:11:18 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Bach",
"Tita Alissa",
""
],
[
"Pedersen",
"Linn",
""
],
[
"Borén†",
"Maria Kinck",
""
],
[
"Temte†",
"Lisa Christoffersen",
""
]
] |
2508.20812
|
Federico Cunico
|
Lorenzo Busellato, Federico Cunico, Diego Dall'Alba, Marco Emporio, Andrea Giachetti, Riccardo Muradore, Marco Cristani
|
Uncertainty Aware-Predictive Control Barrier Functions: Safer Human Robot Interaction through Probabilistic Motion Forecasting
| null | null | null | null |
cs.RO cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
To enable flexible, high-throughput automation in settings where people and robots share workspaces, collaborative robotic cells must reconcile stringent safety guarantees with the need for responsive and effective behavior. A dynamic obstacle is the stochastic, task-dependent variability of human motion: when robots fall back on purely reactive or worst-case envelopes, they brake unnecessarily, stall task progress, and tamper with the fluidity that true Human-Robot Interaction demands. In recent years, learning-based human-motion prediction has rapidly advanced, although most approaches produce worst-case scenario forecasts that often do not treat prediction uncertainty in a well-structured way, resulting in over-conservative planning algorithms, limiting their flexibility. We introduce Uncertainty-Aware Predictive Control Barrier Functions (UA-PCBFs), a unified framework that fuses probabilistic human hand motion forecasting with the formal safety guarantees of Control Barrier Functions. In contrast to other variants, our framework allows for dynamic adjustment of the safety margin thanks to the human motion uncertainty estimation provided by a forecasting module. Thanks to uncertainty estimation, UA-PCBFs empower collaborative robots with a deeper understanding of future human states, facilitating more fluid and intelligent interactions through informed motion planning. We validate UA-PCBFs through comprehensive real-world experiments with an increasing level of realism, including automated setups (to perform exactly repeatable motions) with a robotic hand and direct human-robot interactions (to validate promptness, usability, and human confidence). Relative to state-of-the-art HRI architectures, UA-PCBFs show better performance in task-critical metrics, significantly reducing the number of violations of the robot's safe space during interaction with respect to the state-of-the-art.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:11:26 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Busellato",
"Lorenzo",
""
],
[
"Cunico",
"Federico",
""
],
[
"Dall'Alba",
"Diego",
""
],
[
"Emporio",
"Marco",
""
],
[
"Giachetti",
"Andrea",
""
],
[
"Muradore",
"Riccardo",
""
],
[
"Cristani",
"Marco",
""
]
] |
2508.20813
|
Tao Luo
|
Tao Luo, Han Wu, Tong Yang, Dinggang Shen and Zhiming Cui
|
Adapting Foundation Model for Dental Caries Detection with Dual-View Co-Training
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Accurate dental caries detection from panoramic X-rays plays a pivotal role in preventing lesion progression. However, current detection methods often yield suboptimal accuracy due to subtle contrast variations and diverse lesion morphology of dental caries. In this work, inspired by the clinical workflow where dentists systematically combine whole-image screening with detailed tooth-level inspection, we present DVCTNet, a novel Dual-View Co-Training network for accurate dental caries detection. Our DVCTNet starts with employing automated tooth detection to establish two complementary views: a global view from panoramic X-ray images and a local view from cropped tooth images. We then pretrain two vision foundation models separately on the two views. The global-view foundation model serves as the detection backbone, generating region proposals and global features, while the local-view model extracts detailed features from corresponding cropped tooth patches matched by the region proposals. To effectively integrate information from both views, we introduce a Gated Cross-View Attention (GCV-Atten) module that dynamically fuses dual-view features, enhancing the detection pipeline by integrating the fused features back into the detection model for final caries detection. To rigorously evaluate our DVCTNet, we test it on a public dataset and further validate its performance on a newly curated, high-precision dental caries detection dataset, annotated using both intra-oral images and panoramic X-rays for double verification. Experimental results demonstrate DVCTNet's superior performance against existing state-of-the-art (SOTA) methods on both datasets, indicating the clinical applicability of our method. Our code and labeled dataset are available at https://github.com/ShanghaiTech-IMPACT/DVCTNet.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:13:26 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Luo",
"Tao",
""
],
[
"Wu",
"Han",
""
],
[
"Yang",
"Tong",
""
],
[
"Shen",
"Dinggang",
""
],
[
"Cui",
"Zhiming",
""
]
] |
2508.20814
|
Brandon Alberts
|
Brandon Alberts
|
An Explicit Tauberian Theorem taking Averaged Inputs with an Application to Counting Abelian Number Fields
| null | null | null | null |
math.NT
|
http://creativecommons.org/licenses/by/4.0/
|
Given a Dirichlet series $L(s) = \sum a_n n^{-s}$, the asymptotic growth rate of $\sum_{n\le X} a_n$ can be determined by a Tauberian theorem. Bounds on the error term are typically controlled by the size of $|L(\sigma+it)|$ for fixed real part $\sigma$. We modify this approach to prove new Tauberian theorems with error terms depending only on the average size of $L(\sigma+it)$ as $t$ varies, and we take care to track explicit dependence on various parameters. This often leads to stronger error bounds, and introduces strong connections between asymptotic counting problems and moments of $L$-functions.
We provide self-contained statements of Tauberian theorems in anticipation that these results can be used ``out of the box'' to prove new asymptotic expansions. We demonstrate this by proving square root saving error bounds for the number of $C_n$-extensions of $\mathbb{Q}$ of bounded discriminant when $n=3$, $4$, $8$, $16$, or $2p$ for $p$ an odd prime.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:13:44 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Alberts",
"Brandon",
""
]
] |
2508.20815
|
Chi-Yu Zhou
|
Shiping Liu and Chiyu Zhou
|
Quantitative Obata's theorem in discrete setting
|
18 pages
| null | null | null |
math.DG math.CO math.SP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Under mild assumptions, we show that a connected weighted graph $G$ with lower Ricci curvature bound $K>0$ in the sense of Bakry-\'Emery and the $d$-th non-zero Laplacian eigenvalue $\lambda_d$ close to $K$, with $d$ being the maximal combinatorial vertex degree of $G$, has an underlying combinatorial structure of the $d$-dimensional hypercube graph. Moreover, such a graph $G$ is close in terms of Frobenius distance to a properly weighted hypercube graph. Furthermore, we establish their closeness in terms of eigenfunctions. Our results can be viewed as discrete analogies of the almost rigidity theorem and quantitative Obata's theorem on Rimennian manifolds.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:14:08 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Liu",
"Shiping",
""
],
[
"Zhou",
"Chiyu",
""
]
] |
2508.20816
|
Arthur Gervais
|
Isaac David and Arthur Gervais
|
Multi-Agent Penetration Testing AI for the Web
| null | null | null | null |
cs.CR cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
AI-powered development platforms are making software creation accessible to a broader audience, but this democratization has triggered a scalability crisis in security auditing. With studies showing that up to 40% of AI-generated code contains vulnerabilities, the pace of development now vastly outstrips the capacity for thorough security assessment.
We present MAPTA, a multi-agent system for autonomous web application security assessment that combines large language model orchestration with tool-grounded execution and end-to-end exploit validation. On the 104-challenge XBOW benchmark, MAPTA achieves 76.9% overall success with perfect performance on SSRF and misconfiguration vulnerabilities, 83% success on broken authorization, and strong results on injection attacks including server-side template injection (85%) and SQL injection (83%). Cross-site scripting (57%) and blind SQL injection (0%) remain challenging. Our comprehensive cost analysis across all challenges totals $21.38 with a median cost of $0.073 for successful attempts versus $0.357 for failures. Success correlates strongly with resource efficiency, enabling practical early-stopping thresholds at approximately 40 tool calls or $0.30 per challenge.
MAPTA's real-world findings are impactful given both the popularity of the respective scanned GitHub repositories (8K-70K stars) and MAPTA's low average operating cost of $3.67 per open-source assessment: MAPTA discovered critical vulnerabilities including RCEs, command injections, secret exposure, and arbitrary file write vulnerabilities. Findings are responsibly disclosed, 10 findings are under CVE review.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:14:24 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"David",
"Isaac",
""
],
[
"Gervais",
"Arthur",
""
]
] |
2508.20817
|
Xingchen Zhang
|
He Li, Xinyu Liu, Weihang Kong, Xingchen Zhang
|
FusionCounting: Robust visible-infrared image fusion guided by crowd counting via multi-task learning
|
11 pages, 9 figures
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Most visible and infrared image fusion (VIF) methods focus primarily on optimizing fused image quality. Recent studies have begun incorporating downstream tasks, such as semantic segmentation and object detection, to provide semantic guidance for VIF. However, semantic segmentation requires extensive annotations, while object detection, despite reducing annotation efforts compared with segmentation, faces challenges in highly crowded scenes due to overlapping bounding boxes and occlusion. Moreover, although RGB-T crowd counting has gained increasing attention in recent years, no studies have integrated VIF and crowd counting into a unified framework. To address these challenges, we propose FusionCounting, a novel multi-task learning framework that integrates crowd counting into the VIF process. Crowd counting provides a direct quantitative measure of population density with minimal annotation, making it particularly suitable for dense scenes. Our framework leverages both input images and population density information in a mutually beneficial multi-task design. To accelerate convergence and balance tasks contributions, we introduce a dynamic loss function weighting strategy. Furthermore, we incorporate adversarial training to enhance the robustness of both VIF and crowd counting, improving the model's stability and resilience to adversarial attacks. Experimental results on public datasets demonstrate that FusionCounting not only enhances image fusion quality but also achieves superior crowd counting performance.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:15:18 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Li",
"He",
""
],
[
"Liu",
"Xinyu",
""
],
[
"Kong",
"Weihang",
""
],
[
"Zhang",
"Xingchen",
""
]
] |
2508.20818
|
Anirudh Satheesh
|
Anirudh Satheesh, Keenan Powell, Hua Wei
|
cMALC-D: Contextual Multi-Agent LLM-Guided Curriculum Learning with Diversity-Based Context Blending
|
A shorter version has been accepted to the 2025 Conference on Information and Knowledge Management
| null | null | null |
cs.LG cs.MA
|
http://creativecommons.org/licenses/by/4.0/
|
Many multi-agent reinforcement learning (MARL) algorithms are trained in fixed simulation environments, making them brittle when deployed in real-world scenarios with more complex and uncertain conditions. Contextual MARL (cMARL) addresses this by parameterizing environments with context variables and training a context-agnostic policy that performs well across all environment configurations. Existing cMARL methods attempt to use curriculum learning to help train and evaluate context-agnostic policies, but they often rely on unreliable proxy signals, such as value estimates or generalized advantage estimates that are noisy and unstable in multi-agent settings due to inter-agent dynamics and partial observability. To address these issues, we propose Contextual Multi-Agent LLM-Guided Curriculum Learning with Diversity-Based Context Blending (cMALC-D), a framework that uses Large Language Models (LLMs) to generate semantically meaningful curricula and provide a more robust evaluation signal. To prevent mode collapse and encourage exploration, we introduce a novel diversity-based context blending mechanism that creates new training scenarios by combining features from prior contexts. Experiments in traffic signal control domains demonstrate that cMALC-D significantly improves both generalization and sample efficiency compared to existing curriculum learning baselines. We provide code at https://github.com/DaRL-LibSignal/cMALC-D.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:16:17 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Satheesh",
"Anirudh",
""
],
[
"Powell",
"Keenan",
""
],
[
"Wei",
"Hua",
""
]
] |
2508.20819
|
Luke Postle
|
Michelle Delcourt, Cicely Henderson, Thomas Lesgourgues, Luke Postle
|
Beyond Nash-Williams: Counterexamples to Clique Decomposition Thresholds for All Cliques Larger than Triangles
|
14 pages, 3 figures
| null | null | null |
math.CO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
A central open question in extremal design theory is Nash-Williams' Conjecture from 1970 that every $K_3$-divisible graph on $n$ vertices (for $n$ large enough) with minimum degree at least $3n/4$ has a $K_3$-decomposition. A folklore generalization of Nash-Williams' Conjecture extends this to all $q\ge 4$ by positing that every $K_q$-divisible graph on $n$ vertices (for $n$ large enough) with minimum degree at least $\left(1-\frac{1}{q+1}\right)n$ has a $K_q$-decomposition. We disprove this conjecture for all $q\ge 4$; namely, we show that for each $q\ge 4$, there exists $c > 1$ such that there exist infinitely many $K_q$-divisible graphs $G$ with minimum degree at least $\left(1-\frac{1}{c\cdot(q+1)}\right)v(G)$ and no $K_q$-decomposition; indeed we construct them admitting no fractional $K_q$-decomposition thus disproving the fractional relaxation of this conjecture. Our result also disproves the more general partite version. Indeed, we even show the folklore conjecture is off by a multiplicative factor by showing that for every $\varepsilon > 0$ and every large enough integer $q$, there exist infinitely many $K_q$-divisible graphs $G$ with minimum degree at least $\bigg(1-\frac{1}{\left(\frac{1+\sqrt{2}}{2}-\varepsilon\right)\cdot (q+1)}\bigg)v(G)$ with no (fractional) $K_q$-decomposition.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:17:26 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Delcourt",
"Michelle",
""
],
[
"Henderson",
"Cicely",
""
],
[
"Lesgourgues",
"Thomas",
""
],
[
"Postle",
"Luke",
""
]
] |
2508.20820
|
Baptist Piest
|
Baptist Piest, Jonas B\"ohm, Timoth\'e Estrampes, Annie Pichery, Pawe{\l} Arciszewski, Wolfgang Bartosch, S\"oren Boles, Klaus D\"oringshoff, Michael Elsen, Priyanka Guggilam, Ortwin Hellmig, Christian K\"urbis, Dorthe Leopoldt, Gabriel M\"uller, Alexandros Papakonstantinou, Christian Reichelt, Andr\'e Wenzlawski, Thijs Wendrich, \'Eric Charron, Achim Peters, Klaus Sengstock, Andreas Wicht, Patrick Windpassinger, Jens Grosse, Naceur Gaaloul, Ernst Maria Rasel
|
Apparatus for quantum-mixture research in microgravity
| null | null | null | null |
cond-mat.quant-gas physics.atom-ph quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Experiments with ultracold quantum gases are a rapidly advancing research field with many applications in fundamental physics and quantum technology. Here, we report on a high-flux generation of Bose-Einstein condensate mixtures of $^{41}$K and $^{87}$Rb, using a fully integrated sounding rocket setup. We investigate the release and the free expansion of the quantum mixtures for different orientations to gravity. The release dynamics are governed by the mixture interactions as well as the decaying magnetic field during the release. The latter can be minimized by a dedicated switch-off protocol of the trap generating currents where an exact model enabled us to characterize the interaction effects. Our results establish a new benchmark for generating ultracold mixtures on mobile platforms, with direct relevance for future experiments on interacting quantum gases and tests of the equivalence principle in space.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:17:47 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Piest",
"Baptist",
""
],
[
"Böhm",
"Jonas",
""
],
[
"Estrampes",
"Timothé",
""
],
[
"Pichery",
"Annie",
""
],
[
"Arciszewski",
"Paweł",
""
],
[
"Bartosch",
"Wolfgang",
""
],
[
"Boles",
"Sören",
""
],
[
"Döringshoff",
"Klaus",
""
],
[
"Elsen",
"Michael",
""
],
[
"Guggilam",
"Priyanka",
""
],
[
"Hellmig",
"Ortwin",
""
],
[
"Kürbis",
"Christian",
""
],
[
"Leopoldt",
"Dorthe",
""
],
[
"Müller",
"Gabriel",
""
],
[
"Papakonstantinou",
"Alexandros",
""
],
[
"Reichelt",
"Christian",
""
],
[
"Wenzlawski",
"André",
""
],
[
"Wendrich",
"Thijs",
""
],
[
"Charron",
"Éric",
""
],
[
"Peters",
"Achim",
""
],
[
"Sengstock",
"Klaus",
""
],
[
"Wicht",
"Andreas",
""
],
[
"Windpassinger",
"Patrick",
""
],
[
"Grosse",
"Jens",
""
],
[
"Gaaloul",
"Naceur",
""
],
[
"Rasel",
"Ernst Maria",
""
]
] |
2508.20821
|
Anna Krasznai
|
Anna Krasznai, Sergei Rutkevich and G\'abor Tak\'acs
|
Confinement in the three-state Potts quantum spin chain in extreme ferromagnetic limit
|
44 pages, pdflatex
| null | null | null |
cond-mat.stat-mech hep-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We examine the dynamics of the three-state Potts quantum spin chain in the extreme ferromagnetic regime using perturbation theory in the transverse magnetic field. We demonstrate that this approach provides access to features well beyond the semiclassical method applied previously, including the description of resonant excitations and analytic prediction for the time evolution after a quantum quench. We also demonstrate that it agrees with the meson/bubble spectrum results from exact diagonalisation and the numerical simulations of the time evolution.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:17:51 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Krasznai",
"Anna",
""
],
[
"Rutkevich",
"Sergei",
""
],
[
"Takács",
"Gábor",
""
]
] |
2508.20822
|
Laszlo Gacsi
|
Laszlo Gacsi, Max H. Cohen, and Tamas G. Molnar
|
Activated Backstepping with Control Barrier Functions for the Safe Navigation of Automated Vehicles
|
Accepted to the IEEE Control Systems Letters. 6 pages, 3 figures
| null |
10.1109/LCSYS.2025.3577096
| null |
math.DS math.OC
|
http://creativecommons.org/licenses/by/4.0/
|
This paper introduces a novel safety-critical control method through the synthesis of control barrier functions (CBFs) for systems with high-relative-degree safety constraints. By extending the procedure of CBF backstepping, we propose activated backstepping - a constructive method to synthesize valid CBFs. The novelty of our method is the incorporation of an activation function into the CBF, which offers less conservative safe sets in the state space than standard CBF backstepping. We demonstrate the proposed method on an inverted pendulum example, where we explain the underlying geometric meaning in the state space and provide comparisons with existing CBF synthesis techniques. Finally, we implement our method to achieve collision-free navigation for automated vehicles using a kinematic bicycle model in simulation.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:19:44 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Gacsi",
"Laszlo",
""
],
[
"Cohen",
"Max H.",
""
],
[
"Molnar",
"Tamas G.",
""
]
] |
2508.20823
|
Kang Chen
|
Kang Chen, Yasong Feng, Tianyu Wang
|
Revisit Stochastic Gradient Descent for Strongly Convex Objectives: Tight Uniform-in-Time Bounds
|
12 pages
| null | null | null |
math.OC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Stochastic optimization for strongly convex objectives is a fundamental problem in statistics and optimization. This paper revisits the standard Stochastic Gradient Descent (SGD) algorithm for strongly convex objectives and establishes tight uniform-in-time convergence bounds. We prove that with probability larger than $1 - \beta$, a $\frac{\log \log k + \log (1/\beta)}{k}$ convergence bound simultaneously holds for all $k \in \mathbb{N}_+$, and show that this rate is tight up to constants. Our results also include an improved last-iterate convergence rate for SGD on strongly convex objectives.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:20:49 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Chen",
"Kang",
""
],
[
"Feng",
"Yasong",
""
],
[
"Wang",
"Tianyu",
""
]
] |
2508.20824
|
Yang Gao
|
Yang Gao, Dongjie Wang, Scott Piersall, Ye Zhang, Liqiang Wang
|
GPT-FT: An Efficient Automated Feature Transformation Using GPT for Sequence Reconstruction and Performance Enhancement
|
17 pages, 9 figures. accepted by APWeb-WAIM 2025
| null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Feature transformation plays a critical role in enhancing machine learning model performance by optimizing data representations. Recent state-of-the-art approaches address this task as a continuous embedding optimization problem, converting discrete search into a learnable process. Although effective, these methods often rely on sequential encoder-decoder structures that cause high computational costs and parameter requirements, limiting scalability and efficiency. To address these limitations, we propose a novel framework that accomplishes automated feature transformation through four steps: transformation records collection, embedding space construction with a revised Generative Pre-trained Transformer (GPT) model, gradient-ascent search, and autoregressive reconstruction. In our approach, the revised GPT model serves two primary functions: (a) feature transformation sequence reconstruction and (b) model performance estimation and enhancement for downstream tasks by constructing the embedding space. Such a multi-objective optimization framework reduces parameter size and accelerates transformation processes. Experimental results on benchmark datasets show that the proposed framework matches or exceeds baseline performance, with significant gains in computational efficiency. This work highlights the potential of transformer-based architectures for scalable, high-performance automated feature transformation.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:21:08 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Gao",
"Yang",
""
],
[
"Wang",
"Dongjie",
""
],
[
"Piersall",
"Scott",
""
],
[
"Zhang",
"Ye",
""
],
[
"Wang",
"Liqiang",
""
]
] |
2508.20825
|
Jorik Jooken
|
Jorik Jooken
|
Computer-assisted graph theory: a survey
|
46 pages
| null | null | null |
math.CO cs.DM
|
http://creativecommons.org/licenses/by/4.0/
|
Computers and algorithms play an ever-increasing role in obtaining new results in graph theory. In this survey, we present a broad range of techniques used in computer-assisted graph theory, including the exhaustive generation of all pairwise non-isomorphic graphs within a given class, the use of searchable databases containing graphs and invariants as well as other established and emerging algorithmic paradigms. We cover approaches based on mixed integer linear programming, semidefinite programming, dynamic programming, SAT solving, metaheuristics and machine learning. The techniques are illustrated with numerous detailed results covering several important subareas of graph theory such as extremal graph theory, graph coloring, structural graph theory, spectral graph theory, regular graphs, topological graph theory, special sets in graphs, algebraic graph theory and chemical graph theory. We also present some smaller new results that demonstrate how readily a computer-assisted graph theory approach can be applied once the appropriate tools have been developed.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:21:46 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Jooken",
"Jorik",
""
]
] |
2508.20826
|
Rapha\"el Clou\^atre
|
Rapha\"el Clou\^atre, Poornendu Kumar
|
Annhilating ideals and Agler--McCarthy spectral varieties in the bidisc
|
21 pages
| null | null | null |
math.FA math.CV math.OA
|
http://creativecommons.org/licenses/by/4.0/
|
The closed unit bidisc $\overline{\mathbb{D}}^2$ is known to be a spectral set for any pair $(T_1,T_2)$ of commuting contractions. When each $T_i$ is pure and has finite defect, the pair admits a much smaller spectral set: the closure of a distinguished variety $V$ inside the bidisc $\mathbb{D}^2$. We find conditions on $(T_1,T_2)$ that guarantee that the closure of $V$ is a minimal spectral set. In addition, we examine the relationship between $V$ and the annihilating ideal $\text{Ann}(T_1,T_2)$ in $H^\infty(\mathbb{D}^2)$. While $V$ is typically strictly larger than the zero set of $\text{Ann}(T_1,T_2)$, we isolate a natural constrained isometric co-extension $(S_1,S_2)$ of $(T_1,T_2)$ whose Taylor spectrum is contained in $V$ and is closely linked to the so-called support of $\text{Ann}(T_1,T_2)$. We also characterize when $\text{Ann}(T_1,T_2)$ is the ideal of functions vanishing on the joint point spectrum of $(S_1^*,S_2^*)$.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:22:23 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Clouâtre",
"Raphaël",
""
],
[
"Kumar",
"Poornendu",
""
]
] |
2508.20827
|
Alexei Ivlev
|
A. A. Gavdush, A. V. Ivlev, K. I. Zaytsev, V. E. Ulitko, I. N. Dolganova, S. V. Garnov, B. M. Giuliano, and P. Caselli
|
Broadband spectroscopy of astrophysical ice analogues: III. Scattering properties and porosity of CO and CO$_2$ ices
|
Accepted for publication in A&A
| null | null | null |
astro-ph.IM
|
http://creativecommons.org/licenses/by/4.0/
|
$Context.$ The quantification of the terahertz (THz) and IR optical properties of astrophysical ice analogs, which have different molecular compositions, phases, and structural properties, is required to model both the continuum emission by the dust grains covered with thick icy mantles and the radiative transfer in the dense cold regions of the interstellar medium.
$Aims.$ We developed a model to define a relationship between the THz$-$IR response and the ice porosity. It includes the reduced effective optical properties of porous ices and the additional wave extinction due to scattering on pores. The model is applied to analyze the measured THz$-$IR response of CO and CO$_2$ laboratory ices and to estimate their scattering properties and porosity.
$Methods.$ Our model combines the Bruggeman effective medium theory, the Lorentz-Mie and Rayleigh scattering theories, and the radiative transfer theory to analyze the measured THz$-$IR optical properties of laboratory ices.
$Results.$ We apply this model to show that the electromagnetic-wave scattering in studied laboratory ices occurs mainly in the Rayleigh regime at frequencies below 32 THz. We conclude that pores of different shapes and dimensions can be approximated by spheres of effective radius. By comparing the measured broadband response of our laboratory ices with those of reportedly compact ices from earlier studies, we quantify the scattering properties of our CO and CO$_2$ ice samples. Their porosity is shown to be as high as 15% and 22%, respectively. Underestimating the ice porosity in the data analysis leads to a proportional relative underestimate of the THz$-$IR optical constants.
$Conclusions.$ The scattering properties and porosity of ices have to be quantified along with their THz$-$IR response in order to adequately interpret astrophysical observations.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:22:40 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Gavdush",
"A. A.",
""
],
[
"Ivlev",
"A. V.",
""
],
[
"Zaytsev",
"K. I.",
""
],
[
"Ulitko",
"V. E.",
""
],
[
"Dolganova",
"I. N.",
""
],
[
"Garnov",
"S. V.",
""
],
[
"Giuliano",
"B. M.",
""
],
[
"Caselli",
"P.",
""
]
] |
2508.20828
|
Jie Zhao
|
Jie Zhao, Wanting Ning, Yuxiao Fei, Yubo Feng and Lishuang Li
|
GDLLM: A Global Distance-aware Modeling Approach Based on Large Language Models for Event Temporal Relation Extraction
|
Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP Findings)
| null | null | null |
cs.CL cs.IR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In Natural Language Processing(NLP), Event Temporal Relation Extraction (ETRE) is to recognize the temporal relations of two events. Prior studies have noted the importance of language models for ETRE. However, the restricted pre-trained knowledge of Small Language Models(SLMs) limits their capability to handle minority class relations in imbalanced classification datasets. For Large Language Models(LLMs), researchers adopt manually designed prompts or instructions, which may introduce extra noise, leading to interference with the model's judgment of the long-distance dependencies between events. To address these issues, we propose GDLLM, a Global Distance-aware modeling approach based on LLMs. We first present a distance-aware graph structure utilizing Graph Attention Network(GAT) to assist the LLMs in capturing long-distance dependency features. Additionally, we design a temporal feature learning paradigm based on soft inference to augment the identification of relations with a short-distance proximity band, which supplements the probabilistic information generated by LLMs into the multi-head attention mechanism. Since the global feature can be captured effectively, our framework substantially enhances the performance of minority relation classes and improves the overall learning ability. Experiments on two publicly available datasets, TB-Dense and MATRES, demonstrate that our approach achieves state-of-the-art (SOTA) performance.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:23:39 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Zhao",
"Jie",
""
],
[
"Ning",
"Wanting",
""
],
[
"Fei",
"Yuxiao",
""
],
[
"Feng",
"Yubo",
""
],
[
"Li",
"Lishuang",
""
]
] |
2508.20829
|
Zeyue Zhang
|
Zeyue Zhang, Lin Song, Erkang Bao, Xiaoling Lv, Xinyue Wang
|
ATM-GAD: Adaptive Temporal Motif Graph Anomaly Detection for Financial Transaction Networks
| null | null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Financial fraud detection is essential to safeguard billions of dollars, yet the intertwined entities and fast-changing transaction behaviors in modern financial systems routinely defeat conventional machine learning models. Recent graph-based detectors make headway by representing transactions as networks, but they still overlook two fraud hallmarks rooted in time: (1) temporal motifs--recurring, telltale subgraphs that reveal suspicious money flows as they unfold--and (2) account-specific intervals of anomalous activity, when fraud surfaces only in short bursts unique to each entity. To exploit both signals, we introduce ATM-GAD, an adaptive graph neural network that leverages temporal motifs for financial anomaly detection. A Temporal Motif Extractor condenses each account's transaction history into the most informative motifs, preserving both topology and temporal patterns. These motifs are then analyzed by dual-attention blocks: IntraA reasons over interactions within a single motif, while InterA aggregates evidence across motifs to expose multi-step fraud schemes. In parallel, a differentiable Adaptive Time-Window Learner tailors the observation window for every node, allowing the model to focus precisely on the most revealing time slices. Experiments on four real-world datasets show that ATM-GAD consistently outperforms seven strong anomaly-detection baselines, uncovering fraud patterns missed by earlier methods.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:25:07 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Zhang",
"Zeyue",
""
],
[
"Song",
"Lin",
""
],
[
"Bao",
"Erkang",
""
],
[
"Lv",
"Xiaoling",
""
],
[
"Wang",
"Xinyue",
""
]
] |
2508.20830
|
Krit Duangprom
|
Krit Duangprom, Tryphon Lambrou, and Binod Bhattarai
|
Estimating 2D Keypoints of Surgical Tools Using Vision-Language Models with Low-Rank Adaptation
|
Accepted to MICCAI 2025
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
This paper presents a novel pipeline for 2D keypoint estima- tion of surgical tools by leveraging Vision Language Models (VLMs) fine- tuned using a low rank adjusting (LoRA) technique. Unlike traditional Convolutional Neural Network (CNN) or Transformer-based approaches, which often suffer from overfitting in small-scale medical datasets, our method harnesses the generalization capabilities of pre-trained VLMs. We carefully design prompts to create an instruction-tuning dataset and use them to align visual features with semantic keypoint descriptions. Experimental results show that with only two epochs of fine tuning, the adapted VLM outperforms the baseline models, demonstrating the ef- fectiveness of LoRA in low-resource scenarios. This approach not only improves keypoint detection performance, but also paves the way for future work in 3D surgical hands and tools pose estimation.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:25:32 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Duangprom",
"Krit",
""
],
[
"Lambrou",
"Tryphon",
""
],
[
"Bhattarai",
"Binod",
""
]
] |
2508.20831
|
Rui Chen
|
Rui Chen, Domenico Chiaradia, Antonio Frisoli, Daniele Leonardis
|
A Soft Fabric-Based Thermal Haptic Device for VR and Teleoperation
| null | null | null | null |
cs.RO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper presents a novel fabric-based thermal-haptic interface for virtual reality and teleoperation. It integrates pneumatic actuation and conductive fabric with an innovative ultra-lightweight design, achieving only 2~g for each finger unit. By embedding heating elements within textile pneumatic chambers, the system delivers modulated pressure and thermal stimuli to fingerpads through a fully soft, wearable interface.
Comprehensive characterization demonstrates rapid thermal modulation with heating rates up to 3$^{\circ}$C/s, enabling dynamic thermal feedback for virtual or teleoperation interactions. The pneumatic subsystem generates forces up to 8.93~N at 50~kPa, while optimization of fingerpad-actuator clearance enhances cooling efficiency with minimal force reduction. Experimental validation conducted with two different user studies shows high temperature identification accuracy (0.98 overall) across three thermal levels, and significant manipulation improvements in a virtual pick-and-place tasks. Results show enhanced success rates (88.5\% to 96.4\%, p = 0.029) and improved force control precision (p = 0.013) when haptic feedback is enabled, validating the effectiveness of the integrated thermal-haptic approach for advanced human-machine interaction applications.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:25:35 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Chen",
"Rui",
""
],
[
"Chiaradia",
"Domenico",
""
],
[
"Frisoli",
"Antonio",
""
],
[
"Leonardis",
"Daniele",
""
]
] |
2508.20832
|
Haim Bar
|
Haim Bar and Huyen Nguyen and Joanne Conover
|
A State Model for the Analysis of Stem Cell Proliferation and Differentiation
| null | null | null | null |
stat.AP
|
http://creativecommons.org/licenses/by/4.0/
|
Stem cells are characterized by their ability to self-renew, as well as to differentiate and give rise to new populations of cells. Stem cell divisions are crucial for generative processes that occur during early development, and later in adulthood to support tissue regenerative capabilities. This property of stemness, the ability of self-renewal or tissue-specific differentiation, is also observed in cancer cells facilitating the sustenance of tumor growth, and in bipotent megakaryocytic-erythroid progenitors (MEPs) to produce blood cells. We are interested in modeling the size of the stem cell population required to adequately generate tissues or colonies of cells. We develop a state model that characterizes stem cell divisions and the dynamic changes of the stem cell and differentiated cell populations. In our model, the probabilities of self-renewal and differentiation events that stem cells undergo can vary over time instead of remaining constant throughout the process. We provide an estimation method for the division probabilities and using a simulation study, we show that our method provides good estimates even with a small sample size.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:25:43 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Bar",
"Haim",
""
],
[
"Nguyen",
"Huyen",
""
],
[
"Conover",
"Joanne",
""
]
] |
2508.20833
|
Mark Muldoon
|
Anthony Kwong and Mark Muldoon
|
Modelling birdsong transmission with methods from molecular sequence analysis
|
34 pages, 13 figures
| null | null | null |
q-bio.QM stat.CO
|
http://creativecommons.org/licenses/by/4.0/
|
In many species of songbirds, juvenile males learn their songs from adult male tutors. In this paper we formulate a novel Markov model for birdsong transmission developed by analogy with models used in biological sequence analysis. We fit the model using the recently developed Interacting Particle Langevin Algorithm (IPLA) of Akyildiz et al. (arXiv:2303.13429) and analyse a collection of songs from Java sparrows (Lonchura oryzivora) originally recorded and studied by Masayo Soma and her collaborators. The model proves to have limited predictive power for a number of natural problems associated with song transmission in Java sparrows and we propose reasons for this, including the well-established faithfulness of song-learning and the comparative brevity of Java sparrow songs.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:27:09 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Kwong",
"Anthony",
""
],
[
"Muldoon",
"Mark",
""
]
] |
2508.20834
|
Dominic Hunter Dr
|
Dominic Hunter, Marcin S. Mrozowski, Stuart J. Ingleby, Timothy S. Read, Allan P. McWilliam, James P. McGilligan, Ralf Bauer, Peter D. Schwindt, Paul F. Griffin, and Erling Riis
|
High-Resolution Atomic Magnetometer-Based Imaging of Integrated Circuits and Batteries
|
9 pages, 7 figures
| null | null | null |
physics.atom-ph
|
http://creativecommons.org/licenses/by/4.0/
|
We present a high-resolution magnetic imaging system based on a free-induction-decay (FID) optically pumped magnetometer (OPM), integrated with a two-axis scanning micro-electromechanical systems (MEMS) micromirror for automated beam steering. A double-pass optical geometry enables the millimeter-scale device under test (DUT) to be placed directly behind the vapor cell, minimizing the standoff distance and thereby improving spatial resolution, which is fundamentally limited by atomic spin diffusion and the optical beam waist. The functionality of the imaging system was tested on a diverse set of magnetic sources, including a custom printed-circuit board (PCB) with alternating copper tracks, whose measured results agreed well with theoretical predictions based on the Biot-Savart law. The system successfully resolves spatially varying current distributions, identifies polarity-dependent asymmetries in an integrated bridge rectifier chip, and tracks the dynamic current evolution of a ceramic battery in situ. These capabilities highlight its potential for noninvasive diagnostics of electronic circuits and batteries.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:27:12 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Hunter",
"Dominic",
""
],
[
"Mrozowski",
"Marcin S.",
""
],
[
"Ingleby",
"Stuart J.",
""
],
[
"Read",
"Timothy S.",
""
],
[
"McWilliam",
"Allan P.",
""
],
[
"McGilligan",
"James P.",
""
],
[
"Bauer",
"Ralf",
""
],
[
"Schwindt",
"Peter D.",
""
],
[
"Griffin",
"Paul F.",
""
],
[
"Riis",
"Erling",
""
]
] |
2508.20835
|
Qianyu Zhou
|
Hao Yang, Qianyu Zhou, Haijia Sun, Xiangtai Li, Xuequan Lu, Lizhuang Ma, Shuicheng Yan
|
PointDGRWKV: Generalizing RWKV-like Architecture to Unseen Domains for Point Cloud Classification
| null | null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Domain Generalization (DG) has been recently explored to enhance the generalizability of Point Cloud Classification (PCC) models toward unseen domains. Prior works are based on convolutional networks, Transformer or Mamba architectures, either suffering from limited receptive fields or high computational cost, or insufficient long-range dependency modeling. RWKV, as an emerging architecture, possesses superior linear complexity, global receptive fields, and long-range dependency. In this paper, we present the first work that studies the generalizability of RWKV models in DG PCC. We find that directly applying RWKV to DG PCC encounters two significant challenges: RWKV's fixed direction token shift methods, like Q-Shift, introduce spatial distortions when applied to unstructured point clouds, weakening local geometric modeling and reducing robustness. In addition, the Bi-WKV attention in RWKV amplifies slight cross-domain differences in key distributions through exponential weighting, leading to attention shifts and degraded generalization. To this end, we propose PointDGRWKV, the first RWKV-based framework tailored for DG PCC. It introduces two key modules to enhance spatial modeling and cross-domain robustness, while maintaining RWKV's linear efficiency. In particular, we present Adaptive Geometric Token Shift to model local neighborhood structures to improve geometric context awareness. In addition, Cross-Domain key feature Distribution Alignment is designed to mitigate attention drift by aligning key feature distributions across domains. Extensive experiments on multiple benchmarks demonstrate that PointDGRWKV achieves state-of-the-art performance on DG PCC.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:28:33 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Yang",
"Hao",
""
],
[
"Zhou",
"Qianyu",
""
],
[
"Sun",
"Haijia",
""
],
[
"Li",
"Xiangtai",
""
],
[
"Lu",
"Xuequan",
""
],
[
"Ma",
"Lizhuang",
""
],
[
"Yan",
"Shuicheng",
""
]
] |
2508.20836
|
Ahmed Elgohary
|
Ahmed A. Elgohary, Rohan Palanikumar, Sameh A. Eisa
|
Model-Free Hovering and Source Seeking via Extremum Seeking Control: Experimental Demonstration
| null | null | null | null |
cs.RO math.OC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In a recent effort, we successfully proposed a categorically novel approach to mimic the phenomenoa of hovering and source seeking by flapping insects and hummingbirds using a new extremum seeking control (ESC) approach. Said ESC approach was shown capable of characterizing the physics of hovering and source seeking by flapping systems, providing at the same time uniquely novel opportunity for a model-free, real-time biomimicry control design. In this paper, we experimentally test and verify, for the first time in the literature, the potential of ESC in flapping robots to achieve model-free, real-time controlled hovering and source seeking. The results of this paper, while being restricted to 1D, confirm the premise of introducing ESC as a natural control method and biomimicry mechanism to the field of flapping flight and robotics.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:29:07 GMT"
}
] | 2025-08-29T00:00:00 |
[
[
"Elgohary",
"Ahmed A.",
""
],
[
"Palanikumar",
"Rohan",
""
],
[
"Eisa",
"Sameh A.",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.