id
string | submitter
string | authors
string | title
string | comments
string | journal-ref
string | doi
string | report-no
string | categories
string | license
string | abstract
string | versions
list | update_date
timestamp[s] | authors_parsed
list |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2505.07361 | Dennis Mayer | Dennis Mayer, Evgenii Titov, Fabiano Lever, Lisa Mehner, Marta L. Murillo-S\'anchez, Constantin Walz, John Bozek, Peter Saalfrank, Markus G\"uhr | Valence photoelectron spectra of thiouracils in the gas phase | 42 pages, 17 figures | null | null | null | physics.chem-ph | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We present a combined experimental and theoretical study of the vibrationally resolved valence photoelectron spectra of the complete series of thiouracils (2-thiouracil, 4-thiouracil and 2,4-dithiouracil) for binding energies between 8 and 17 eV. The theoretical spectra were calculated using equation-of-motion coupled cluster theory for ionization potential (EOM-IP-CCSD) combined with the time-independent double-harmonic adiabatic Hessian approach. For all three thiouracils, the first ionization potential is found between 8.4 and 8.7 eV, which is 1 eV lower than for the canonical nucleobase uracil. Ionization bands up to 12 eV show strong vibrational progressions and are well reproduced by the calculations. These bands are attributed to the ionization of (primarily) sulfur- and oxygen-localized valence molecular orbitals. For higher binding energies, the calculations indicate that nonadiabatic couplings are important for the interpretation of the photoelectron spectra. | [
{
"version": "v1",
"created": "Mon, 12 May 2025 08:55:12 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 11:41:11 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Mayer",
"Dennis",
""
],
[
"Titov",
"Evgenii",
""
],
[
"Lever",
"Fabiano",
""
],
[
"Mehner",
"Lisa",
""
],
[
"Murillo-Sánchez",
"Marta L.",
""
],
[
"Walz",
"Constantin",
""
],
[
"Bozek",
"John",
""
],
[
"Saalfrank",
"Peter",
""
],
[
"Gühr",
"Markus",
""
]
] |
2505.07635 | Dazhuo Qiu | Dazhuo Qiu, Haolai Che, Arijit Khan, Yinghui Wu | Interpreting Graph Inference with Skyline Explanations | null | null | null | null | cs.LG cs.DB | http://creativecommons.org/licenses/by/4.0/ | Inference queries have been routinely issued to graph machine learning models such as graph neural networks (GNNs) for various network analytical tasks. Nevertheless, GNNs outputs are often hard to interpret comprehensively. Existing methods typically compromise to individual pre-defined explainability measures (such as fidelity), which often leads to biased, ``one-sided'' interpretations. This paper introduces skyline explanation, a new paradigm that interprets GNN output by simultaneously optimizing multiple explainability measures of users' interests. (1) We propose skyline explanations as a Pareto set of explanatory subgraphs that dominate others over multiple explanatory measures. We formulate skyline explanation as a multi-criteria optimization problem, and establish its hardness results. (2) We design efficient algorithms with an onion-peeling approach, which strategically prioritizes nodes and removes unpromising edges to incrementally assemble skyline explanations. (3) We also develop an algorithm to diversify the skyline explanations to enrich the comprehensive interpretation. (4) We introduce efficient parallel algorithms with load-balancing strategies to scale skyline explanation for large-scale GNN-based inference. Using real-world and synthetic graphs, we experimentally verify our algorithms' effectiveness and scalability. | [
{
"version": "v1",
"created": "Mon, 12 May 2025 15:05:46 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 13:40:49 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Qiu",
"Dazhuo",
""
],
[
"Che",
"Haolai",
""
],
[
"Khan",
"Arijit",
""
],
[
"Wu",
"Yinghui",
""
]
] |
2505.08601 | Jinchi Zhu | Jinchi Zhu, Zhou Zhao, Hailong Lei, Xiaoguang Wang, Jialiang Lu, Jing Li, Qianqian Tang, Jiachen Shen, Gui-Song Xia, Bo Du and Yongchao Xu | Rejoining fragmented ancient bamboo slips with physics-driven deep learning | null | null | null | null | cs.CV cond-mat.mtrl-sci | http://creativecommons.org/licenses/by/4.0/ | Bamboo slips are a crucial medium for recording ancient civilizations in East Asia, and offers invaluable archaeological insights for reconstructing the Silk Road, studying material culture exchanges, and global history. However, many excavated bamboo slips have been fragmented into thousands of irregular pieces, making their rejoining a vital yet challenging step for understanding their content. Here we introduce WisePanda, a physics-driven deep learning framework designed to rejoin fragmented bamboo slips. Based on the physics of fracture and material deterioration, WisePanda automatically generates synthetic training data that captures the physical properties of bamboo fragmentations. This approach enables the training of a matching network without requiring manually paired samples, providing ranked suggestions to facilitate the rejoining process. Compared to the leading curve matching method, WisePanda increases Top-50 matching accuracy from 36% to 52% among more than one thousand candidate fragments. Archaeologists using WisePanda have experienced substantial efficiency improvements (approximately 20 times faster) when rejoining fragmented bamboo slips. This research demonstrates that incorporating physical principles into deep learning models can significantly enhance their performance, transforming how archaeologists restore and study fragmented artifacts. WisePanda provides a new paradigm for addressing data scarcity in ancient artifact restoration through physics-driven machine learning. | [
{
"version": "v1",
"created": "Tue, 13 May 2025 14:16:53 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 02:44:45 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Zhu",
"Jinchi",
""
],
[
"Zhao",
"Zhou",
""
],
[
"Lei",
"Hailong",
""
],
[
"Wang",
"Xiaoguang",
""
],
[
"Lu",
"Jialiang",
""
],
[
"Li",
"Jing",
""
],
[
"Tang",
"Qianqian",
""
],
[
"Shen",
"Jiachen",
""
],
[
"Xia",
"Gui-Song",
""
],
[
"Du",
"Bo",
""
],
[
"Xu",
"Yongchao",
""
]
] |
2505.09497 | Sourav Pal | Sourav Pal, Abhik Basu | Are shocks in conserved driven single-file motions with bottlenecks universal? | Preliminary version, comments are welcome | null | null | null | cond-mat.stat-mech | http://creativecommons.org/licenses/by/4.0/ | Driven single-file motion sets the paradigm for wide variety of one-dimensional directed movements, ranging from intracellular transport and urban traffic to ant trails and controlled robot swarms. Motivated by the phenomenologies of these systems in closed geometries, regulated by number conservation and bottlenecks, we explore the domain walls (DWs) or shocks in a conceptual one-dimensional cellular automaton with a fixed particle number and a bottleneck. For high entry and exit rates of the cellular automaton, and with {sufficiently large} particle numbers, the DWs formed are independent of the associated rate parameters, revealing a {\em hitherto unknown universality} in their {\em shapes}. In contrast, the DWs do depend upon these parameters, if they are small, and hence have nonuniversal shapes. Our predictions on the DWs are testable in experiments. | [
{
"version": "v1",
"created": "Wed, 14 May 2025 15:44:20 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 14:54:08 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Pal",
"Sourav",
""
],
[
"Basu",
"Abhik",
""
]
] |
2505.13112 | Rodrigo Maulen-Soto | Rodrigo Maulen-Soto (SU, LPSM (UMR\_8001)), Claire Boyer (IUF), Pierre Marion (EPFL) | Attention-based clustering | null | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Transformers have emerged as a powerful neural network architecture capable of tackling a wide range of learning tasks. In this work, we provide a theoretical analysis of their ability to automatically extract structure from data in an unsupervised setting. In particular, we demonstrate their suitability for clustering when the input data is generated from a Gaussian mixture model. To this end, we study a simplified two-head attention layer and define a population risk whose minimization with unlabeled data drives the head parameters to align with the true mixture centroids. This phenomenon highlights the ability of attention-based layers to capture underlying distributional structure. We further examine an attention layer with key, query, and value matrices fixed to the identity, and show that, even without any trainable parameters, it can perform in-context quantization, revealing the surprising capacity of transformer-based methods to adapt dynamically to input-specific distributions. | [
{
"version": "v1",
"created": "Mon, 19 May 2025 13:39:56 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 08:04:46 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Maulen-Soto",
"Rodrigo",
"",
"SU, LPSM"
],
[
"Boyer",
"Claire",
"",
"IUF"
],
[
"Marion",
"Pierre",
"",
"EPFL"
]
] |
2505.13886 | Jingqi Tong | Jingqi Tong, Jixin Tang, Hangcheng Li, Yurong Mou, Ming Zhang, Jun Zhao, Yanbo Wen, Fan Song, Jiahao Zhan, Yuyang Lu, Chaoran Tao, Zhiyuan Guo, Jizhou Yu, Tianhao Cheng, Changhao Jiang, Zhen Wang, Tao Liang, Zhihui Fei, Mingyang Wan, Guojun Ma, Weifeng Ge, Guanhua Chen, Tao Gui, Xipeng Qiu, Qi Zhang, Xuanjing Huang | Code2Logic: Game-Code-Driven Data Synthesis for Enhancing VLMs General Reasoning | 63 pages, 23 figures, submitted to NeurIPS 2025 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Visual-language Chain-of-Thought (CoT) data resources are relatively scarce compared to text-only counterparts, limiting the improvement of reasoning capabilities in Vision Language Models (VLMs). However, high-quality vision-language reasoning data is expensive and labor-intensive to annotate. To address this issue, we leverage a promising resource: game code, which naturally contains logical structures and state transition processes. Therefore, we propose Code2Logic, a novel game-code-driven approach for multimodal reasoning data synthesis. Our approach leverages Large Language Models (LLMs) to adapt game code, enabling automatic acquisition of reasoning processes and results through code execution. Using the Code2Logic approach, we developed the GameQA dataset to train and evaluate VLMs. GameQA is cost-effective and scalable, offers controllable difficulty gradation and is diverse with 30 games and 158 tasks. Surprisingly, despite training solely on game data, VLMs demonstrated out of domain generalization, specifically Qwen2.5-VL-7B improving performance by 2.33% across 7 diverse vision-language benchmarks. Our code, dataset and models are available at https://github.com/tongjingqi/Code2Logic. | [
{
"version": "v1",
"created": "Tue, 20 May 2025 03:47:44 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 14:30:25 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Tong",
"Jingqi",
""
],
[
"Tang",
"Jixin",
""
],
[
"Li",
"Hangcheng",
""
],
[
"Mou",
"Yurong",
""
],
[
"Zhang",
"Ming",
""
],
[
"Zhao",
"Jun",
""
],
[
"Wen",
"Yanbo",
""
],
[
"Song",
"Fan",
""
],
[
"Zhan",
"Jiahao",
""
],
[
"Lu",
"Yuyang",
""
],
[
"Tao",
"Chaoran",
""
],
[
"Guo",
"Zhiyuan",
""
],
[
"Yu",
"Jizhou",
""
],
[
"Cheng",
"Tianhao",
""
],
[
"Jiang",
"Changhao",
""
],
[
"Wang",
"Zhen",
""
],
[
"Liang",
"Tao",
""
],
[
"Fei",
"Zhihui",
""
],
[
"Wan",
"Mingyang",
""
],
[
"Ma",
"Guojun",
""
],
[
"Ge",
"Weifeng",
""
],
[
"Chen",
"Guanhua",
""
],
[
"Gui",
"Tao",
""
],
[
"Qiu",
"Xipeng",
""
],
[
"Zhang",
"Qi",
""
],
[
"Huang",
"Xuanjing",
""
]
] |
2505.14281 | Victor Falgas-Ravry | Victor Falgas-Ravry, Adva Mond, Rik Sarkar, Victor Souza | On problems in extremal multigraph theory | 40 pages, 2 figures. Minor changes and edits from v1 | null | null | null | math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A multigraph G is said to be an (s,q)-graph if every s-set of vertices in G supports at most q edges (counting multiplicities). In this paper we consider the maximal sum and product of edge multiplicities in an (s,q)-graph on n vertices. These are multigraph analogues of a problem of Erd\H{o}s raised by F\"uredi and K\"undgen and Mubayi and Terry respectively, with applications to counting problems and extremal hypergraph theory.
We make major progress, settling conjectures of Day, Falgas-Ravry and Treglown and of Falgas-Ravry, establishing intricate behaviour for both the sum and the product problems, and providing both a general picture and evidence that the problems may prove computationally intractable in general. | [
{
"version": "v1",
"created": "Tue, 20 May 2025 12:32:08 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 07:58:52 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Falgas-Ravry",
"Victor",
""
],
[
"Mond",
"Adva",
""
],
[
"Sarkar",
"Rik",
""
],
[
"Souza",
"Victor",
""
]
] |
2505.14588 | David Byrne | Martin Baily, David Byrne, Aidan Kane, and Paul Soto | Generative AI at the Crossroads: Light Bulb, Dynamo, or Microscope? | null | null | null | null | econ.GN q-fin.EC | http://creativecommons.org/licenses/by-nc-nd/4.0/ | With the advent of generative AI (genAI), the potential scope of artificial intelligence has increased dramatically, but the future effect of genAI on productivity remains uncertain. The effect of the technology on the innovation process is a crucial open question. Some inventions, such as the light bulb, temporarily raise productivity growth as adoption spreads, but the effect fades when the market is saturated; that is, the level of output per hour is permanently higher but the growth rate is not. In contrast, two types of technologies stand out as having longer-lived effects on productivity growth. First, there are technologies known as general-purpose technologies (GPTs). GPTs (1) are widely adopted, (2) spur abundant knock-on innovations (new goods and services, process efficiencies, and business reorganization), and (3) show continual improvement, refreshing this innovation cycle; the electric dynamo is an example. Second, there are inventions of methods of invention (IMIs). IMIs increase the efficiency of the research and development process via improvements to observation, analysis, communication, or organization; the compound microscope is an example. We show that GenAI has the characteristics of both a GPT and an IMI -- an encouraging sign that genAI will raise the \textit{level} of productivity. Even so, genAI's contribution to productivity \textit{growth} will depend on the speed with which that level is attained and, historically, integrating revolutionary technologies into the economy is a protracted process. | [
{
"version": "v1",
"created": "Tue, 20 May 2025 16:40:51 GMT"
},
{
"version": "v2",
"created": "Mon, 16 Jun 2025 14:19:00 GMT"
},
{
"version": "v3",
"created": "Wed, 02 Jul 2025 18:40:55 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Baily",
"Martin",
""
],
[
"Byrne",
"David",
""
],
[
"Kane",
"Aidan",
""
],
[
"Soto",
"Paul",
""
]
] |
2505.15053 | Tatsuya Abe | Tatsuya Abe | Owicki--Gries Logic for Timestamp Semantics | null | null | null | null | cs.LO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Whereas an extension with non-interference of Hoare logic for sequential programs Owicki--Gries logic ensures the correctness of concurrent programs on strict consistency, it is unsound to weak memory models adopted by modern computer architectures and specifications of programming languages. This paper proposes a novel non-interference notion and provides concurrent program logic sound to timestamp semantics corresponding to a weak memory model that allows delays in the effects of store instructions. This paper reports three theoretically interesting techniques for modifying non-interference to support delays in the effects of store instructions. The techniques contribute to a better understanding of constructing concurrent program logic. | [
{
"version": "v1",
"created": "Wed, 21 May 2025 03:20:16 GMT"
},
{
"version": "v2",
"created": "Sun, 25 May 2025 11:01:54 GMT"
},
{
"version": "v3",
"created": "Wed, 04 Jun 2025 00:37:33 GMT"
},
{
"version": "v4",
"created": "Thu, 03 Jul 2025 00:46:49 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Abe",
"Tatsuya",
""
]
] |
2505.15057 | Frederic Wang | Frederic Wang, Jonathan I. Tamir | Non-rigid Motion Correction for MRI Reconstruction via Coarse-To-Fine Diffusion Models | ICIP 2025 | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by/4.0/ | Magnetic Resonance Imaging (MRI) is highly susceptible to motion artifacts due to the extended acquisition times required for k-space sampling. These artifacts can compromise diagnostic utility, particularly for dynamic imaging. We propose a novel alternating minimization framework that leverages a bespoke diffusion model to jointly reconstruct and correct non-rigid motion-corrupted k-space data. The diffusion model uses a coarse-to-fine denoising strategy to capture large overall motion and reconstruct the lower frequencies of the image first, providing a better inductive bias for motion estimation than that of standard diffusion models. We demonstrate the performance of our approach on both real-world cine cardiac MRI datasets and complex simulated rigid and non-rigid deformations, even when each motion state is undersampled by a factor of 64x. Additionally, our method is agnostic to sampling patterns, anatomical variations, and MRI scanning protocols, as long as some low frequency components are sampled during each motion state. | [
{
"version": "v1",
"created": "Wed, 21 May 2025 03:27:21 GMT"
},
{
"version": "v2",
"created": "Thu, 29 May 2025 06:38:03 GMT"
},
{
"version": "v3",
"created": "Thu, 03 Jul 2025 04:39:51 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Wang",
"Frederic",
""
],
[
"Tamir",
"Jonathan I.",
""
]
] |
2505.15075 | Hao Wang | Hao Wang, Pinzhi Huang, Jihan Yang, Saining Xie, Daisuke Kawahara | Traveling Across Languages: Benchmarking Cross-Lingual Consistency in Multimodal LLMs | https://github.com/nlp-waseda/traveling-across-languages | null | null | null | cs.CL cs.AI cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The rapid evolution of multimodal large language models (MLLMs) has significantly enhanced their real-world applications. However, achieving consistent performance across languages, especially when integrating cultural knowledge, remains a significant challenge. To better assess this issue, we introduce two new benchmarks: KnowRecall and VisRecall, which evaluate cross-lingual consistency in MLLMs. KnowRecall is a visual question answering benchmark designed to measure factual knowledge consistency in 15 languages, focusing on cultural and historical questions about global landmarks. VisRecall assesses visual memory consistency by asking models to describe landmark appearances in 9 languages without access to images. Experimental results reveal that state-of-the-art MLLMs, including proprietary ones, still struggle to achieve cross-lingual consistency. This underscores the need for more robust approaches that produce truly multilingual and culturally aware models. | [
{
"version": "v1",
"created": "Wed, 21 May 2025 03:43:37 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 10:35:35 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Wang",
"Hao",
""
],
[
"Huang",
"Pinzhi",
""
],
[
"Yang",
"Jihan",
""
],
[
"Xie",
"Saining",
""
],
[
"Kawahara",
"Daisuke",
""
]
] |
2505.15470 | David Navarro-Giron\'es | D. Navarro-Giron\'es, M. Crocce, E. Gazta\~naga, A. Wittje, M. Siudek, H. Hoekstra, H. Hildebrandt, B. Joachimi, R. Paviot, C.M. Baugh, J. Carretero, R. Casas, F.J. Castander, M. Eriksen, E. Fernandez, P. Fosalba, J. Garc\'ia-Bellido, R. Miquel, C. Padilla, P. Renard, E. S\'anchez, S. Serrano, I. Sevilla-Noarbe, P. Tallada-Cresp\'i | The PAU Survey: Measuring intrinsic galaxy alignments in deep wide fields as a function of colour, luminosity, stellar mass and redshift | 30 pages, 24 figures, submitted to MNRAS. New version: Figures 9 and 12 updated | null | null | null | astro-ph.CO astro-ph.GA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present the measurements and constraints of intrinsic alignments (IA) in the Physics of the Accelerating Universe Survey (PAUS) deep wide fields, which include the W1 and W3 fields from the Canada-France-Hawaii Telescope Legacy Survey (CFHTLS) and the G09 field from the Kilo-Degree Survey (KiDS). Our analyses cover 51deg$^{2}$, in the photometric redshift (photo-$z$) range $0.1 < z_{\mathrm{b}} < 1$ and a magnitude limit $i_{\mathrm{AB}}<22$. The precise photo-$z$s and the luminosity coverage of PAUS enable robust IA measurements, which are key for setting informative priors for upcoming stage-IV surveys. For red galaxies, we detect an increase in IA amplitude with both luminosity and stellar mass, extending previous results towards fainter and less massive regimes. As a function of redshift, we observe strong IA signals at intermediate ($z_{\mathrm{b}}\sim0.55$) and high ($z_{\mathrm{b}}\sim0.75$) redshift bins. However, we find no significant trend of IA evolution with redshift after accounting for the varying luminosities across redshift bins, consistent with the literature. For blue galaxies, no significant IA signal is detected, with $A_{1}=0.68_{-0.51}^{+0.53}$ when splitting only by galaxy colour, yielding some of the tightest constraints to date for the blue population and constraining a regime of very faint and low-mass galaxies. | [
{
"version": "v1",
"created": "Wed, 21 May 2025 12:46:20 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 18:09:37 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Navarro-Gironés",
"D.",
""
],
[
"Crocce",
"M.",
""
],
[
"Gaztañaga",
"E.",
""
],
[
"Wittje",
"A.",
""
],
[
"Siudek",
"M.",
""
],
[
"Hoekstra",
"H.",
""
],
[
"Hildebrandt",
"H.",
""
],
[
"Joachimi",
"B.",
""
],
[
"Paviot",
"R.",
""
],
[
"Baugh",
"C. M.",
""
],
[
"Carretero",
"J.",
""
],
[
"Casas",
"R.",
""
],
[
"Castander",
"F. J.",
""
],
[
"Eriksen",
"M.",
""
],
[
"Fernandez",
"E.",
""
],
[
"Fosalba",
"P.",
""
],
[
"García-Bellido",
"J.",
""
],
[
"Miquel",
"R.",
""
],
[
"Padilla",
"C.",
""
],
[
"Renard",
"P.",
""
],
[
"Sánchez",
"E.",
""
],
[
"Serrano",
"S.",
""
],
[
"Sevilla-Noarbe",
"I.",
""
],
[
"Tallada-Crespí",
"P.",
""
]
] |
2505.15621 | Shuyin Ouyang | Shuyin Ouyang, Dong Huang, Jingwen Guo, Zeyu Sun, Qihao Zhu, Jie M. Zhang | DSCodeBench: A Realistic Benchmark for Data Science Code Generation | null | null | null | null | cs.SE | http://creativecommons.org/publicdomain/zero/1.0/ | We introduce DSCodeBench, a new benchmark designed to evaluate large language models (LLMs) on complicated and realistic data science code generation tasks. DSCodeBench consists of 1,000 carefully constructed problems sourced from realistic problems from GitHub across ten widely used Python data science libraries. Compared to the current state-of-the-art benchmark DS-1000, DSCodeBench offers a more challenging and representative testbed, longer code solutions, more comprehensive data science libraries, clearer and better structured problem descriptions, and stronger test suites. To construct the DSCodeBench, we develop a robust pipeline that combines task scope selection, code construction, test case generation, and problem description synthesis. The process is paired with rigorous manual editing to ensure alignment and enhance evaluation reliability. Experimental result shows that DSCodeBench exhibits robust scaling behavior, where larger models systematically outperform smaller ones, validating its ability to distinguish model capabilities. The best LLM we test, GPT-4o, has a pass@1 of 0.202, indicating that LLMs still have a large room to improve for realistic data science code generation tasks. We believe DSCodeBench will serve as a rigorous and trustworthy foundation for advancing LLM-based data science programming. | [
{
"version": "v1",
"created": "Wed, 21 May 2025 15:11:26 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 20:43:52 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Ouyang",
"Shuyin",
""
],
[
"Huang",
"Dong",
""
],
[
"Guo",
"Jingwen",
""
],
[
"Sun",
"Zeyu",
""
],
[
"Zhu",
"Qihao",
""
],
[
"Zhang",
"Jie M.",
""
]
] |
2505.15976 | Marco Olivieri | Marco Olivieri | A Lee-Huang-Yang type expansion for the thermodynamic energy density of a dilute mixture of Bose gases | Corrected some typos and small mistakes which do not affect the result. Added a comment on the class of potentials satisfying the assumptions of the main theorem | null | null | null | math-ph math.AP math.MP math.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider a dilute gas in 3D composed of two species of bosons interacting through positive inter-species and intra-species pairwise potentials. We prove a second order expansion for the energy density in the thermodynamic limit. For the case of compactly supported, integrable potentials, we derive the correct second order of the expansion. If we make the further assumption of having soft potentials, we also derive the correct coefficient of the second order and the resulting formula is coherent with the physics literature. If we let the density and scattering length of one of the species go to zero, we obtain the Lee-Huang-Yang formula for one species of bosons. The paper also contains a proof of BEC for a mixture of bosons in a box with length scale larger than the Gross-Pitaevskii one. | [
{
"version": "v1",
"created": "Wed, 21 May 2025 19:52:16 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 13:24:49 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Olivieri",
"Marco",
""
]
] |
2505.16341 | Yaxin Hou | Yaxin Hou, Yuheng Jia | A Square Peg in a Square Hole: Meta-Expert for Long-Tailed Semi-Supervised Learning | The paper is accepted by ICML 2025 | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper studies the long-tailed semi-supervised learning (LTSSL) with distribution mismatch, where the class distribution of the labeled training data follows a long-tailed distribution and mismatches with that of the unlabeled training data. Most existing methods introduce auxiliary classifiers (experts) to model various unlabeled data distributions and produce pseudo-labels, but the expertises of various experts are not fully utilized. We observe that different experts are good at predicting different intervals of samples, e.g., long-tailed expert is skilled in samples located in the head interval and uniform expert excels in samples located in the medium interval. Therefore, we propose a dynamic expert assignment module that can estimate the class membership (i.e., head, medium, or tail class) of samples, and dynamically assigns suitable expert to each sample based on the estimated membership to produce high-quality pseudo-label in the training phase and produce prediction in the testing phase. We also theoretically reveal that integrating different experts' strengths will lead to a smaller generalization error bound. Moreover, we find that the deeper features are more biased toward the head class but with more discriminative ability, while the shallower features are less biased but also with less discriminative ability. We, therefore, propose a multi-depth feature fusion module to utilize different depth features to mitigate the model bias. Our method demonstrates its effectiveness through comprehensive experiments on the CIFAR-10-LT, STL-10-LT, and SVHN-LT datasets across various settings. The code is available at https://github.com/yaxinhou/Meta-Expert. | [
{
"version": "v1",
"created": "Thu, 22 May 2025 07:56:44 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 11:12:29 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Hou",
"Yaxin",
""
],
[
"Jia",
"Yuheng",
""
]
] |
2505.17514 | Sergei V. Ketov | Daniel Frolovsky and Sergei V. Ketov | Are single-field models of inflation and PBH production ruled out by ACT observations? | 9 pages, 4 figures, LaTeX; misprints corrected, more comments and references added | null | null | IPMU25-0026 | astro-ph.CO gr-qc | http://creativecommons.org/licenses/by/4.0/ | The data release from the Atacama Cosmology Telescope (ACT) imposes stronger constraints on primordial black hole (PBH) formation in single-field inflation models than the Planck data. In particular, the updated Cosmic Microwave Background (CMB) radiation measurements favour a $higher$ scalar spectral index $n_s$ and its $positive$ running $\alpha_s$. Even in the absence of PBH production, the updated constraints together place many single-field models under tension with the CMB data. To explore this tension, we study PBH formation in an $\alpha$-attractor E-model. We investigate an impact of bending of the inflaton potential plateau toward reconciling the model with the new bounds for CMB observables. We find that attempts to increase $n_s$ by upward bending lead to negative values of $\alpha_s$ that are disfavoured by the new bounds. A possible model-building approach to resolve the tension is proposed. | [
{
"version": "v1",
"created": "Fri, 23 May 2025 06:08:55 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 11:58:06 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Frolovsky",
"Daniel",
""
],
[
"Ketov",
"Sergei V.",
""
]
] |
2505.18422 | Saleh Afroogh | Saleh Afroogh, Kush R. Varshney, Jason D'Cruz | A Task-Driven Human-AI Collaboration: When to Automate, When to Collaborate, When to Challenge | null | null | null | null | cs.CY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | According to several empirical investigations, despite enhancing human capabilities, human-AI cooperation frequently falls short of expectations and fails to reach true synergy. We propose a task-driven framework that reverses prevalent approaches by assigning AI roles according to how the task's requirements align with the capabilities of AI technology. Three major AI roles are identified through task analysis across risk and complexity dimensions: autonomous, assistive/collaborative, and adversarial. We show how proper human-AI integration maintains meaningful agency while improving performance by methodically mapping these roles to various task types based on current empirical findings. This framework lays the foundation for practically effective and morally sound human-AI collaboration that unleashes human potential by aligning task attributes to AI capabilities. It also provides structured guidance for context-sensitive automation that complements human strengths rather than replacing human judgment. | [
{
"version": "v1",
"created": "Fri, 23 May 2025 23:19:15 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 02:30:11 GMT"
},
{
"version": "v3",
"created": "Wed, 28 May 2025 18:22:17 GMT"
},
{
"version": "v4",
"created": "Thu, 03 Jul 2025 13:59:14 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Afroogh",
"Saleh",
""
],
[
"Varshney",
"Kush R.",
""
],
[
"D'Cruz",
"Jason",
""
]
] |
2505.20127 | Lior Limonad | Fabiana Fournier, Lior Limonad, Yuval David | Agentic AI Process Observability: Discovering Behavioral Variability | 12 pages, 7 figures | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | AI agents that leverage Large Language Models (LLMs) are increasingly becoming core building blocks of modern software systems. A wide range of frameworks is now available to support the specification of such applications. These frameworks enable the definition of agent setups using natural language prompting, which specifies the roles, goals, and tools assigned to the various agents involved. Within such setups, agent behavior is non-deterministic for any given input, highlighting the critical need for robust debugging and observability tools. In this work, we explore the use of process and causal discovery applied to agent execution trajectories as a means of enhancing developer observability. This approach aids in monitoring and understanding the emergent variability in agent behavior. Additionally, we complement this with LLM-based static analysis techniques to distinguish between intended and unintended behavioral variability. We argue that such instrumentation is essential for giving developers greater control over evolving specifications and for identifying aspects of functionality that may require more precise and explicit definitions. | [
{
"version": "v1",
"created": "Mon, 26 May 2025 15:26:07 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 10:05:46 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Fournier",
"Fabiana",
""
],
[
"Limonad",
"Lior",
""
],
[
"David",
"Yuval",
""
]
] |
2505.20410 | Toby Wiseman | Xiaoyi Liu, Harvey S. Reall, Jorge E. Santos, Toby Wiseman | Ill-posedness of the Cauchy problem for linearized gravity in a cavity with conformal boundary conditions | 33 pages, 2 figures; v2 minor additions | null | null | null | gr-qc hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider Lorentzian General Relativity in a cavity with a timelike boundary, with conformal boundary conditions and also a generalization of these boundary conditions. We focus on the linearized gravitational dynamics about the static empty cavity whose boundary has spherical spatial geometry. It has been recently shown that there exist dynamical instabilities, whose angular dependence is given in terms of spherical harmonics $Y_{\ell m}$, and whose coefficient of exponential growth in time goes as $\sim \ell^{1/3}$. We use these modes to construct a sequence of solutions for which the initial data converge to zero as $\ell \rightarrow \infty$ but for which the solution itself does not converge to zero. This implies a lack of continuity of solutions on initial data, which shows that the initial value problem with these boundary conditions is not well-posed. This is in tension with recent mathematical work on well-posedness for such boundary conditions. | [
{
"version": "v1",
"created": "Mon, 26 May 2025 18:00:52 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 15:52:52 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Liu",
"Xiaoyi",
""
],
[
"Reall",
"Harvey S.",
""
],
[
"Santos",
"Jorge E.",
""
],
[
"Wiseman",
"Toby",
""
]
] |
2505.20697 | Zachary Brown | Zachary C. Brown and David Carlson | Generating Hypotheses of Dynamic Causal Graphs in Neuroscience: Leveraging Generative Factor Models of Observed Time Series | null | null | null | null | cs.LG cs.AI stat.AP stat.ML | http://creativecommons.org/licenses/by/4.0/ | The field of hypothesis generation promises to reduce costs in neuroscience by narrowing the range of interventional studies needed to study various phenomena. Existing machine learning methods can generate scientific hypotheses from complex datasets, but many approaches assume causal relationships are static over time, limiting their applicability to systems with dynamic, state-dependent behavior, such as the brain. While some techniques attempt dynamic causal discovery through factor models, they often restrict relationships to linear patterns or impose other simplifying assumptions. We propose a novel method that models dynamic graphs as a conditionally weighted superposition of static graphs, where each static graph can capture nonlinear relationships. This approach enables the detection of complex, time-varying interactions between variables beyond linear limitations. Our method improves f1-scores of predicted dynamic causal patterns by roughly 22-28% on average over baselines in some of our experiments, with some improvements reaching well over 60%. A case study on real brain data demonstrates our method's ability to uncover relationships linked to specific behavioral states, offering valuable insights into neural dynamics. | [
{
"version": "v1",
"created": "Tue, 27 May 2025 04:06:47 GMT"
},
{
"version": "v2",
"created": "Fri, 30 May 2025 14:13:47 GMT"
},
{
"version": "v3",
"created": "Wed, 02 Jul 2025 20:42:02 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Brown",
"Zachary C.",
""
],
[
"Carlson",
"David",
""
]
] |
2505.21290 | Alan Frieze | Colin Cooper, Alan Frieze | Rainbow copies of spanning subgraphs | There is an error in the proof | null | null | null | math.CO cs.DM | http://creativecommons.org/licenses/by/4.0/ | Let $G_{n,p}^{[\kappa]}$ denote the space of $n$-vertex edge coloured graphs, where each edge occurs independently with probability $p$. The colour of each existing edge is chosen independently and uniformly at random from the set $[\kappa]$. We consider the threshold for the existence of rainbow colored copies of a spanning subgraph $H$. We provide lower bounds on $p$ and $\kappa$ sufficient to prove the existence of such copies w.h.p. | [
{
"version": "v1",
"created": "Tue, 27 May 2025 14:54:32 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 00:47:13 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Cooper",
"Colin",
""
],
[
"Frieze",
"Alan",
""
]
] |
2505.21303 | Liang-Bi Wu | Libo Xie, Liang-Bi Wu and Zong-Kuan Guo | Spectrum instability and greybody factor stability for parabolic approximation of Regge-Wheeler potential | 15 pages, 7 figures and 2 tables. This version has been accepted for publication as a regular article in PRD | null | null | null | gr-qc | http://creativecommons.org/publicdomain/zero/1.0/ | We investigate the stability of QNM spectra and greybody factors in the Schwarzschild black hole by approximating the Regge-Wheeler potential with a piecewise parabolic form and treating the deviation as a perturbation. We find that QNM spectra are sensitive to small perturbations, while greybody factors remain stable. This piecewise parabolic approximated potential gives rise to the long-lived modes whose imaginary parts remain close to zero and decrease slowly with overtone number increasing. The reflection coefficient shows distinct resonance feature in the high-frequency regime that are absent in the original R-W case. For the calculation of greybody factors, we employ an analytic method based on transfer matrix technique, and this approach can also be effectively used in other effective potential cases. | [
{
"version": "v1",
"created": "Tue, 27 May 2025 15:05:21 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 13:24:54 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Xie",
"Libo",
""
],
[
"Wu",
"Liang-Bi",
""
],
[
"Guo",
"Zong-Kuan",
""
]
] |
2505.21880 | Yu-Ming Chang | Yu-Lun Song, Chung-En Tsern, Che-Cheng Wu, Yu-Ming Chang, Syuan-Bo Huang, Wei-Chu Chen, Michael Chia-Liang Lin, Yu-Ta Lin | Incorporating LLMs for Large-Scale Urban Complex Mobility Simulation | 8 pages, 8 figures. This paper is reviewed and accepted by the CUPUM (Computational Urban Planning and Urban Management) Conference held by University College London (UCL) in 2025 | null | 10.17605/OSF.IO/ABYQH | null | cs.MA cs.AI cs.CL cs.CY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This study presents an innovative approach to urban mobility simulation by integrating a Large Language Model (LLM) with Agent-Based Modeling (ABM). Unlike traditional rule-based ABM, the proposed framework leverages LLM to enhance agent diversity and realism by generating synthetic population profiles, allocating routine and occasional locations, and simulating personalized routes. Using real-world data, the simulation models individual behaviors and large-scale mobility patterns in Taipei City. Key insights, such as route heat maps and mode-specific indicators, provide urban planners with actionable information for policy-making. Future work focuses on establishing robust validation frameworks to ensure accuracy and reliability in urban planning applications. | [
{
"version": "v1",
"created": "Wed, 28 May 2025 01:54:28 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 06:38:34 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Song",
"Yu-Lun",
""
],
[
"Tsern",
"Chung-En",
""
],
[
"Wu",
"Che-Cheng",
""
],
[
"Chang",
"Yu-Ming",
""
],
[
"Huang",
"Syuan-Bo",
""
],
[
"Chen",
"Wei-Chu",
""
],
[
"Lin",
"Michael Chia-Liang",
""
],
[
"Lin",
"Yu-Ta",
""
]
] |
2505.21900 | Hyukpyo Hong | Hyukpyo Hong, Diego Rojas La Luz, Gheorghe Craciun | Ubiquitous Asymptotic Robustness in Biochemical Systems | This include two files: a main text and Supplementary Information. 17 pages, 4 figures, 2 tables for the main text; 29 pages, 1 figure, 18 tables for the Supplementary Information | null | null | null | math.DS q-bio.QM | http://creativecommons.org/licenses/by/4.0/ | Living systems maintain stable internal states despite environmental fluctuations. Absolute concentration robustness (ACR) is a striking homeostatic phenomenon in which the steady-state concentration of a molecular species remains invariant to changes in total molecular supply. Although experimental studies have reported approximate-but not exact-robustness in steady-state concentrations, such behavior has often been attributed to exact ACR motifs perturbed by measurement noise or minor side reactions, rather than recognized as a structural property of the network itself. In this work, we highlight a previously underappreciated phenomenon, which we term asymptotic ACR (aACR): approximate robustness can emerge solely from the architecture of the reaction network, without requiring parameters being negligible or the presence of an exact ACR motif. We find that aACR is far more common than classical ACR, as demonstrated in systems such as the Escherichia coli EnvZ-OmpR system and MAPK signaling cascade. Furthermore, we mathematically prove that such ubiquity stems solely from network structure. Finally, we reveal a counterintuitive feature of aACR in systems with multiple conserved quantities, revealing subtle distinctions in how robustness manifests in complex biochemical networks. | [
{
"version": "v1",
"created": "Wed, 28 May 2025 02:27:34 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 23:54:58 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Hong",
"Hyukpyo",
""
],
[
"La Luz",
"Diego Rojas",
""
],
[
"Craciun",
"Gheorghe",
""
]
] |
2505.22343 | Zhonghao Lyu | Zhonghao Lyu, Yulan Gao, Junting Chen, Hongyang Du, Jie Xu, Kaibin Huang, and Dong In Kim | Empowering Intelligent Low-altitude Economy with Large AI Model Deployment | null | null | null | null | eess.SP cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Low-altitude economy (LAE) represents an emerging economic paradigm that redefines commercial and social aerial activities. Large artificial intelligence models (LAIMs) offer transformative potential to further enhance the intelligence of LAE services. However, deploying LAIMs in LAE poses several challenges, including the significant gap between their computational/storage demands and the limited onboard resources of LAE entities, the mismatch between lab-trained LAIMs and dynamic physical environments, and the inefficiencies of traditional decoupled designs for sensing, communication, and computation. To address these issues, we first propose a hierarchical system architecture tailored for LAIM deployment and present representative LAE application scenarios. Next, we explore key enabling techniques that facilitate the mutual co-evolution of LAIMs and low-altitude systems, and introduce a task-oriented execution pipeline for scalable and adaptive service delivery. Then, the proposed framework is validated through real-world case studies. Finally, we outline open challenges to inspire future research. | [
{
"version": "v1",
"created": "Wed, 28 May 2025 13:27:07 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 14:03:18 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Lyu",
"Zhonghao",
""
],
[
"Gao",
"Yulan",
""
],
[
"Chen",
"Junting",
""
],
[
"Du",
"Hongyang",
""
],
[
"Xu",
"Jie",
""
],
[
"Huang",
"Kaibin",
""
],
[
"Kim",
"Dong In",
""
]
] |
2505.22502 | Dominic Lowe | Dominic Lowe, M.S. Kim, Roberto Bondesan | Assessing Quantum Advantage for Gaussian Process Regression | 18 pages, 2 figures. Version 2 contains updated figures and a slightly revised discussion for additional clarity | null | null | null | quant-ph cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Gaussian Process Regression is a well-known machine learning technique for which several quantum algorithms have been proposed. We show here that in a wide range of scenarios these algorithms show no exponential speedup. We achieve this by rigorously proving that the condition number of a kernel matrix scales at least linearly with the matrix size under general assumptions on the data and kernel. We additionally prove that the sparsity and Frobenius norm of a kernel matrix scale linearly under similar assumptions. The implications for the quantum algorithms runtime are independent of the complexity of loading classical data on a quantum computer and also apply to dequantised algorithms. We supplement our theoretical analysis with numerical verification for popular kernels in machine learning. | [
{
"version": "v1",
"created": "Wed, 28 May 2025 15:50:56 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 11:27:35 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Lowe",
"Dominic",
""
],
[
"Kim",
"M. S.",
""
],
[
"Bondesan",
"Roberto",
""
]
] |
2505.22618 | Chengyue Wu | Chengyue Wu, Hao Zhang, Shuchen Xue, Zhijian Liu, Shizhe Diao, Ligeng Zhu, Ping Luo, Song Han, Enze Xie | Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Diffusion-based large language models (Diffusion LLMs) have shown promise for non-autoregressive text generation with parallel decoding capabilities. However, the practical inference speed of open-sourced Diffusion LLMs often lags behind autoregressive models due to the lack of Key-Value (KV) Cache and quality degradation when decoding multiple tokens simultaneously. To bridge this gap, we introduce a novel block-wise approximate KV Cache mechanism tailored for bidirectional diffusion models, enabling cache reuse with negligible performance drop. Additionally, we identify the root cause of generation quality degradation in parallel decoding as the disruption of token dependencies under the conditional independence assumption. To address this, we propose a confidence-aware parallel decoding strategy that selectively decodes tokens exceeding a confidence threshold, mitigating dependency violations and maintaining generation quality. Experimental results on LLaDA and Dream models across multiple LLM benchmarks demonstrate up to \textbf{27.6$\times$ throughput} improvement with minimal accuracy loss, closing the performance gap with autoregressive models and paving the way for practical deployment of Diffusion LLMs. | [
{
"version": "v1",
"created": "Wed, 28 May 2025 17:39:15 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 05:11:54 GMT"
},
{
"version": "v3",
"created": "Thu, 03 Jul 2025 04:51:05 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Wu",
"Chengyue",
""
],
[
"Zhang",
"Hao",
""
],
[
"Xue",
"Shuchen",
""
],
[
"Liu",
"Zhijian",
""
],
[
"Diao",
"Shizhe",
""
],
[
"Zhu",
"Ligeng",
""
],
[
"Luo",
"Ping",
""
],
[
"Han",
"Song",
""
],
[
"Xie",
"Enze",
""
]
] |
2505.22768 | Mert Cakiroglu | Mert Onur Cakiroglu, Idil Bilge Altun, Mehmet Dalkilic, Elham Buxton and Hasan Kurban | Multivariate de Bruijn Graphs: A Symbolic Graph Framework for Time Series Forecasting | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Time series forecasting remains a challenging task for foundation models due to temporal heterogeneity, high dimensionality, and the lack of inherent symbolic structure. In this work, we propose DRAGON (Discrete Representation and Augmented Graph encoding Over de BruijN Graphs), a novel encoder that introduces Multivariate de Bruijn Graphs (MdBGs) to bridge the gap between symbolic representations and neural modeling. DRAGON discretizes continuous input sequences and maps them onto a fixed graph structure, enabling dynamic context recovery via graph-based attention. Integrated as an auxiliary module within a dual-branch architecture, DRAGON augments conventional CNN-based encoders with symbolic, structure-aware representations. All code developed for this study is available at: https://github.com/KurbanIntelligenceLab/MultdBG-Time-Series-Library | [
{
"version": "v1",
"created": "Wed, 28 May 2025 18:36:26 GMT"
},
{
"version": "v2",
"created": "Sat, 28 Jun 2025 04:29:45 GMT"
},
{
"version": "v3",
"created": "Thu, 03 Jul 2025 00:58:49 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Cakiroglu",
"Mert Onur",
""
],
[
"Altun",
"Idil Bilge",
""
],
[
"Dalkilic",
"Mehmet",
""
],
[
"Buxton",
"Elham",
""
],
[
"Kurban",
"Hasan",
""
]
] |
2505.22868 | Md Hasibul Amin | Md Hasibul Amin, Mohammadreza Mohammadi, Jason D. Bakos, and Ramtin Zand | CrossNAS: A Cross-Layer Neural Architecture Search Framework for PIM Systems | null | Proceedings of the Great Lakes Symposium on VLSI 2025 | 10.1145/3716368.3735178 | null | cs.ET cs.AR cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we propose the CrossNAS framework, an automated approach for exploring a vast, multidimensional search space that spans various design abstraction layers-circuits, architecture, and systems-to optimize the deployment of machine learning workloads on analog processing-in-memory (PIM) systems. CrossNAS leverages the single-path one-shot weight-sharing strategy combined with the evolutionary search for the first time in the context of PIM system mapping and optimization. CrossNAS sets a new benchmark for PIM neural architecture search (NAS), outperforming previous methods in both accuracy and energy efficiency while maintaining comparable or shorter search times. | [
{
"version": "v1",
"created": "Wed, 28 May 2025 21:00:49 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Amin",
"Md Hasibul",
""
],
[
"Mohammadi",
"Mohammadreza",
""
],
[
"Bakos",
"Jason D.",
""
],
[
"Zand",
"Ramtin",
""
]
] |
2505.23115 | Liu Yicheng | Yunshen Wang, Yicheng Liu, Tianyuan Yuan, Yingshi Liang, Xiuyu Yang, Honggang Zhang, Hang Zhao | Diffusion-Based Generative Models for 3D Occupancy Prediction in Autonomous Driving | ICRA 2025 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Accurately predicting 3D occupancy grids from visual inputs is critical for autonomous driving, but current discriminative methods struggle with noisy data, incomplete observations, and the complex structures inherent in 3D scenes. In this work, we reframe 3D occupancy prediction as a generative modeling task using diffusion models, which learn the underlying data distribution and incorporate 3D scene priors. This approach enhances prediction consistency, noise robustness, and better handles the intricacies of 3D spatial structures. Our extensive experiments show that diffusion-based generative models outperform state-of-the-art discriminative approaches, delivering more realistic and accurate occupancy predictions, especially in occluded or low-visibility regions. Moreover, the improved predictions significantly benefit downstream planning tasks, highlighting the practical advantages of our method for real-world autonomous driving applications. | [
{
"version": "v1",
"created": "Thu, 29 May 2025 05:34:22 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 06:55:33 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Wang",
"Yunshen",
""
],
[
"Liu",
"Yicheng",
""
],
[
"Yuan",
"Tianyuan",
""
],
[
"Liang",
"Yingshi",
""
],
[
"Yang",
"Xiuyu",
""
],
[
"Zhang",
"Honggang",
""
],
[
"Zhao",
"Hang",
""
]
] |
2505.23486 | Lun Du | Ke Weng, Lun Du, Sirui Li, Wangyue Lu, Haozhe Sun, Hengyu Liu, Tiancheng Zhang | Autoformalization in the Era of Large Language Models: A Survey | null | null | null | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Autoformalization, the process of transforming informal mathematical propositions into verifiable formal representations, is a foundational task in automated theorem proving, offering a new perspective on the use of mathematics in both theoretical and applied domains. Driven by the rapid progress in artificial intelligence, particularly large language models (LLMs), this field has witnessed substantial growth, bringing both new opportunities and unique challenges. In this survey, we provide a comprehensive overview of recent advances in autoformalization from both mathematical and LLM-centric perspectives. We examine how autoformalization is applied across various mathematical domains and levels of difficulty, and analyze the end-to-end workflow from data preprocessing to model design and evaluation. We further explore the emerging role of autoformalization in enhancing the verifiability of LLM-generated outputs, highlighting its potential to improve both the trustworthiness and reasoning capabilities of LLMs. Finally, we summarize key open-source models and datasets supporting current research, and discuss open challenges and promising future directions for the field. | [
{
"version": "v1",
"created": "Thu, 29 May 2025 14:34:54 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 06:03:07 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Weng",
"Ke",
""
],
[
"Du",
"Lun",
""
],
[
"Li",
"Sirui",
""
],
[
"Lu",
"Wangyue",
""
],
[
"Sun",
"Haozhe",
""
],
[
"Liu",
"Hengyu",
""
],
[
"Zhang",
"Tiancheng",
""
]
] |
2505.23772 | William Buchanan Prof | William J Buchanan and Jamie Gilchrist | Anamorphic Cryptography using Baby-Step Giant-Step Recovery | null | null | null | null | cs.CR | http://creativecommons.org/licenses/by/4.0/ | In 2022, Persianom, Phan and Yung outlined the creation of Anamorphic Cryptography. With this, we can create a public key to encrypt data, and then have two secret keys. These secret keys are used to decrypt the cipher into different messages. So, one secret key is given to the Dictator (who must be able to decrypt all the messages), and the other is given to Alice. Alice can then decrypt the ciphertext to a secret message that the Dictator cannot see. This paper outlines the implementation of Anamorphic Cryptography using ECC (Elliptic Curve Cryptography), such as with the secp256k1 curve. This gives considerable performance improvements over discrete logarithm-based methods with regard to security for a particular bit length. Overall, it outlines how the secret message sent to Alice is hidden within the random nonce value, which is used within the encryption process, and which is cancelled out when the Dictator decrypts the ciphertext. It also shows that the BSGS (Baby-step Giant-step) variant significantly outperforms unoptimised elliptic curve methods. | [
{
"version": "v1",
"created": "Mon, 21 Apr 2025 12:53:32 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 14:52:27 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Buchanan",
"William J",
""
],
[
"Gilchrist",
"Jamie",
""
]
] |
2506.00612 | Wenlong Deng | Running Yang, Wenlong Deng, Minghui Chen, Yuyin Zhou, Xiaoxiao Li | Enhancing Clinical Multiple-Choice Questions Benchmarks with Knowledge Graph Guided Distractor Generation | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Clinical tasks such as diagnosis and treatment require strong decision-making abilities, highlighting the importance of rigorous evaluation benchmarks to assess the reliability of large language models (LLMs). In this work, we introduce a knowledge-guided data augmentation framework that enhances the difficulty of clinical multiple-choice question (MCQ) datasets by generating distractors (i.e., incorrect choices that are similar to the correct one and may confuse existing LLMs). Using our KG-based pipeline, the generated choices are both clinically plausible and deliberately misleading. Our approach involves multi-step, semantically informed walks on a medical knowledge graph to identify distractor paths-associations that are medically relevant but factually incorrect-which then guide the LLM in crafting more deceptive distractors. We apply the designed knowledge graph guided distractor generation (KGGDG) pipline, to six widely used medical QA benchmarks and show that it consistently reduces the accuracy of state-of-the-art LLMs. These findings establish KGGDG as a powerful tool for enabling more robust and diagnostic evaluations of medical LLMs. | [
{
"version": "v1",
"created": "Sat, 31 May 2025 15:51:09 GMT"
},
{
"version": "v2",
"created": "Tue, 03 Jun 2025 05:28:26 GMT"
},
{
"version": "v3",
"created": "Thu, 03 Jul 2025 16:50:12 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Yang",
"Running",
""
],
[
"Deng",
"Wenlong",
""
],
[
"Chen",
"Minghui",
""
],
[
"Zhou",
"Yuyin",
""
],
[
"Li",
"Xiaoxiao",
""
]
] |
2506.01631 | Yanjie Zhao | Zehao Wu, Yanjie Zhao, Haoyu Wang | Gradient-Based Model Fingerprinting for LLM Similarity Detection and Family Classification | null | null | null | null | cs.LG cs.AI cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As Large Language Models (LLMs) become integral software components in modern applications, unauthorized model derivations through fine-tuning, merging, and redistribution have emerged as critical software engineering challenges. Unlike traditional software where clone detection and license compliance are well-established, the LLM ecosystem lacks effective mechanisms to detect model lineage and enforce licensing agreements. This gap is particularly problematic when open-source model creators, such as Meta's LLaMA, require derivative works to maintain naming conventions for attribution, yet no technical means exist to verify compliance.
To fill this gap, treating LLMs as software artifacts requiring provenance tracking, we present TensorGuard, a gradient-based fingerprinting framework for LLM similarity detection and family classification. Our approach extracts model-intrinsic behavioral signatures by analyzing gradient responses to random input perturbations across tensor layers, operating independently of training data, watermarks, or specific model formats. TensorGuard supports the widely-adopted safetensors format and constructs high-dimensional fingerprints through statistical analysis of gradient features. These fingerprints enable two complementary capabilities: direct pairwise similarity assessment between arbitrary models through distance computation, and systematic family classification of unknown models via the K-Means clustering algorithm with domain-informed centroid initialization using known base models. Experimental evaluation on 58 models comprising 8 base models and 50 derivatives across five model families (Llama, Qwen, Gemma, Phi, Mistral) demonstrates 94% classification accuracy under our centroid-initialized K-Means clustering. | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 13:08:01 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 15:36:23 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Wu",
"Zehao",
""
],
[
"Zhao",
"Yanjie",
""
],
[
"Wang",
"Haoyu",
""
]
] |
2506.01768 | Peng Guo | Peng Guo and Frank X. Lee | Toward extracting scattering phase shift from integrated correlation functions IV: Coulomb corrections | typos are corrected | null | null | null | hep-lat cond-mat.other cond-mat.quant-gas hep-ph nucl-th | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The formalism developed in Refs.~\cite{Guo:2023ecc,Guo:2024zal,Guo:2024pvt} that relates the integrated correlation functions for a trapped system to the infinite volume scattering phase shifts through a weighted integral is further extended to include Coulomb interaction between charged particles. The original formalism cannot be applied due to different divergent asymptotic behavior resulting from the long-range nature of the Coulomb force. We show that a modified formula in which the difference of integrated correlation functions between particles interacting with Coulomb plus short-range interaction and with Coulomb interaction alone is free of divergence, and has rapid approach to its infinite volume limit. Using an exactly solvable model, we demonstrate that the short-range potential scattering phase shifts can be reliably extracted from the formula in the presence of Coulomb interaction. | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 15:15:26 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 18:04:05 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Guo",
"Peng",
""
],
[
"Lee",
"Frank X.",
""
]
] |
2506.01772 | Simon-Raphael Fischer | Simon-Raphael Fischer | Extension of algebroids Part I: The Construction | v2: fixing some typos and wrong labelling; v3: improving proof of Thm. 2.1 such that it is independent of the reference | null | null | null | math.DG | http://creativecommons.org/licenses/by/4.0/ | In this series of two papers we will generalise the concept of extending a Lie algebroid by a Lie algebra bundle, leading to a notion of extending a Lie algebroid by another Lie algebroid whose orbits lie in the orbits of the former algebroid. The resulting Lie algebroid's anchor will be the sum of the two initial anchors such that the constructions will be similar to matched pairs of Lie algebroids, but with the major difference that we will allow curvatures. In this part of this series we will focus on the canonical construction making use of strict covariant adjustments, a generalisation of Maurer-Cartan forms in the context of gauge theories equipped with a Lie groupoid action instead of a Lie group action. That is, a Cartan connection with certain conditions on the curvature. The second paper will introduce and explain the obstruction of the extension provided here. Examples will include locally split structures as in Poisson geometry.
As a side result we achieve strong hints towards a possible obstruction theory for certain Cartan connections on Lie algebroids, which will be related to the obstruction of (non-trivial) action algebroids; generalising the statement of the action algebroid structure induced by flat Cartan connections. | [
{
"version": "v1",
"created": "Mon, 02 Jun 2025 15:18:09 GMT"
},
{
"version": "v2",
"created": "Mon, 16 Jun 2025 18:30:04 GMT"
},
{
"version": "v3",
"created": "Thu, 03 Jul 2025 15:35:34 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Fischer",
"Simon-Raphael",
""
]
] |
2506.02453 | Kunyu Wang | Kunyu Wang, Xueyang Fu, Yuanfei Bao, Chengjie Ge, Chengzhi Cao, Wei Zhai, Zheng-Jun Zha | PAID: Pairwise Angular-Invariant Decomposition for Continual Test-Time Adaptation | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Continual Test-Time Adaptation (CTTA) aims to online adapt a pre-trained model to changing environments during inference. Most existing methods focus on exploiting target data, while overlooking another crucial source of information, the pre-trained weights, which encode underutilized domain-invariant priors. This paper takes the geometric attributes of pre-trained weights as a starting point, systematically analyzing three key components: magnitude, absolute angle, and pairwise angular structure. We find that the pairwise angular structure remains stable across diverse corrupted domains and encodes domain-invariant semantic information, suggesting it should be preserved during adaptation. Based on this insight, we propose PAID (Pairwise Angular-Invariant Decomposition), a prior-driven CTTA method that decomposes weight into magnitude and direction, and introduces a learnable orthogonal matrix via Householder reflections to globally rotate direction while preserving the pairwise angular structure. During adaptation, only the magnitudes and the orthogonal matrices are updated. PAID achieves consistent improvements over recent SOTA methods on four widely used CTTA benchmarks, demonstrating that preserving pairwise angular structure offers a simple yet effective principle for CTTA. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 05:18:15 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 06:22:46 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Wang",
"Kunyu",
""
],
[
"Fu",
"Xueyang",
""
],
[
"Bao",
"Yuanfei",
""
],
[
"Ge",
"Chengjie",
""
],
[
"Cao",
"Chengzhi",
""
],
[
"Zhai",
"Wei",
""
],
[
"Zha",
"Zheng-Jun",
""
]
] |
2506.02801 | Juan Carlos Buitrago Oropeza | Juan Carlos Buitrago Oropeza | Concentration of the maximum size of an induced subtree in moderately sparse random graphs | 27 pages | null | null | null | math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Kamaldinov, Skorkin, and Zhukovskii proved that the maximum size of an induced subtree in the binomial random graph $G(n,p)$ is concentrated at two consecutive points, whenever $p\in(0,1)$ is a constant. Using improved bounds on the second moment of the number of induced subtrees, we show that the same result holds when $n^{-\frac{e-2}{3e-2}+\varepsilon}\leq p=o(1)$. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 12:32:26 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 08:28:32 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Oropeza",
"Juan Carlos Buitrago",
""
]
] |
2506.02825 | Vince Lyzinski | Tong Qi, Vera Andersson, Peter Viechnicki, Vince Lyzinski | Asymptotically perfect seeded graph matching without edge correlation (and applications to inference) | 10 figures, 35 pages | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present the OmniMatch algorithm for seeded multiple graph matching. In the setting of $d$-dimensional Random Dot Product Graphs (RDPG), we prove that under mild assumptions, OmniMatch with $s$ seeds asymptotically and efficiently perfectly aligns $O(s^{\alpha})$ unseeded vertices -- for $\alpha<2\wedge d/4$ -- across multiple networks even in the presence of no edge correlation. We demonstrate the effectiveness of our algorithm across numerous simulations and in the context of shuffled graph hypothesis testing. In the shuffled testing setting, testing power is lost due to the misalignment/shuffling of vertices across graphs, and we demonstrate the capacity of OmniMatch to correct for misaligned vertices prior to testing and hence recover the lost testing power. We further demonstrate the algorithm on a pair of data examples from connectomics and machine translation. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 12:54:24 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 19:19:46 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Qi",
"Tong",
""
],
[
"Andersson",
"Vera",
""
],
[
"Viechnicki",
"Peter",
""
],
[
"Lyzinski",
"Vince",
""
]
] |
2506.03764 | R\'ois\'in Luo | R\'ois\'in Luo, James McDermott, Colm O'Riordan | Higher-Order Singular-Value Derivatives of Rectangular Real Matrices | Adding authors; change title; change template | null | null | null | stat.ML cs.LG | http://creativecommons.org/licenses/by/4.0/ | We present a theoretical framework for deriving the general $n$-th order Fr\'echet derivatives of singular values in real rectangular matrices, by leveraging reduced resolvent operators from Kato's analytic perturbation theory for self-adjoint operators. Deriving closed-form expressions for higher-order derivatives of singular values is notoriously challenging through standard matrix-analysis techniques. To overcome this, we treat a real rectangular matrix as a compact operator on a finite-dimensional Hilbert space, and embed the rectangular matrix into a block self-adjoint operator so that non-symmetric perturbations are captured. Applying Kato's asymptotic eigenvalue expansion to this construction, we obtain a general, closed-form expression for the infinitesimal $n$-th order spectral variations. Specializing to $n=2$ and deploying on a Kronecker-product representation with matrix convention yield the Hessian of a singular value, not found in literature. By bridging abstract operator-theoretic perturbation theory with matrices, our framework equips researchers with a practical toolkit for higher-order spectral sensitivity studies in random matrix applications (e.g., adversarial perturbation in deep learning). | [
{
"version": "v1",
"created": "Wed, 04 Jun 2025 09:28:35 GMT"
},
{
"version": "v2",
"created": "Thu, 05 Jun 2025 07:40:31 GMT"
},
{
"version": "v3",
"created": "Thu, 03 Jul 2025 11:36:31 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Luo",
"Róisín",
""
],
[
"McDermott",
"James",
""
],
[
"O'Riordan",
"Colm",
""
]
] |
2506.03997 | Laura Giordano | Mario Alviano, Laura Giordano and Daniele Theseider Dupr\'e | A framework for Conditional Reasoning in Answer Set Programming | 23 pages; version v1 has been accepted for publication as a Technical Communication at ICLP 2025 | null | null | null | cs.AI cs.LO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper we introduce a Conditional Answer Set Programming framework (Conditional ASP) for the definition of conditional extensions of Answer Set Programming (ASP). The approach builds on a conditional logic with typicality, and on the combination of a conditional knowledge base with an ASP program, and allows for conditional reasoning over the answer sets of the program. The formalism relies on a multi-preferential semantics (and on the KLM preferential semantics, as a special case) to provide an interpretation of conditionals. | [
{
"version": "v1",
"created": "Wed, 04 Jun 2025 14:25:34 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 12:50:18 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Alviano",
"Mario",
""
],
[
"Giordano",
"Laura",
""
],
[
"Dupré",
"Daniele Theseider",
""
]
] |
2506.05547 | Fabio Natali | B.S. Lonardoni and F. Natali | On the orbital stability of periodic snoidal waves for the $\phi^4-$equation | 20 pages | null | null | null | math.AP | http://creativecommons.org/licenses/by/4.0/ | The main purpose of this paper is to investigate the global well-posedness and orbital stability of odd periodic traveling waves for the $\phi^4$-equation in the Sobolev space of periodic functions with zero mean. We establish new results on the global well-posedness of weak solutions by combining a semigroup approach with energy estimates. As a consequence, we prove the orbital stability of odd periodic waves by applying a Morse index theorem to the constrained linearized operator defined in the Sobolev space with the zero mean property. | [
{
"version": "v1",
"created": "Thu, 05 Jun 2025 19:50:52 GMT"
},
{
"version": "v2",
"created": "Wed, 18 Jun 2025 12:25:52 GMT"
},
{
"version": "v3",
"created": "Thu, 03 Jul 2025 13:40:30 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Lonardoni",
"B. S.",
""
],
[
"Natali",
"F.",
""
]
] |
2506.06004 | T. Reid Alderson | {\DJ}esika Kolari\'c, Chi Fung Willis Chow, Rita Zi Zhu, Agnes Toth-Petroczy, T. Reid Alderson, Iva Priti\v{s}anac | Into the Unknown: From Structure to Disorder in Protein Function Prediction | 3 Figures, 2 Boxes, 1 Table, 1 Glossary, 5k words | null | null | null | q-bio.BM | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Intrinsically disordered regions (IDRs) account for one-third of the human proteome and play essential biological roles. However, predicting the functions of IDRs remains a major challenge due to their lack of stable structures, rapid sequence evolution, and context-dependent behavior. Many predictors of protein function neglect or underperform on IDRs. Recent advances in computational biology and machine learning, including protein language models, alignment-free approaches, and IDR-specific methods, have revealed conserved bulk features and local motifs within IDRs that are linked to function. This review highlights emerging computational methods that map the sequence-function relationship in IDRs, outlines critical challenges in IDR function annotation, and proposes a community-driven framework to accelerate interpretable functional predictions for IDRs. | [
{
"version": "v1",
"created": "Fri, 06 Jun 2025 11:49:21 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 12:32:18 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Kolarić",
"Đesika",
""
],
[
"Chow",
"Chi Fung Willis",
""
],
[
"Zhu",
"Rita Zi",
""
],
[
"Toth-Petroczy",
"Agnes",
""
],
[
"Alderson",
"T. Reid",
""
],
[
"Pritišanac",
"Iva",
""
]
] |
2506.06231 | Mohammad Jalali | Mohammad Jalali, Bahar Dibaei Nia, Farzan Farnia | Towards an Explainable Comparison and Alignment of Feature Embeddings | null | null | null | null | cs.LG cs.AI cs.CV math.SP | http://creativecommons.org/licenses/by/4.0/ | While several feature embedding models have been developed in the literature, comparisons of these embeddings have largely focused on their numerical performance in classification-related downstream applications. However, an interpretable comparison of different embeddings requires identifying and analyzing mismatches between sample groups clustered within the embedding spaces. In this work, we propose the \emph{Spectral Pairwise Embedding Comparison (SPEC)} framework to compare embeddings and identify their differences in clustering a reference dataset. Our approach examines the kernel matrices derived from two embeddings and leverages the eigendecomposition of the difference kernel matrix to detect sample clusters that are captured differently by the two embeddings. We present a scalable implementation of this kernel-based approach, with computational complexity that grows linearly with the sample size. Furthermore, we introduce an optimization problem using this framework to align two embeddings, ensuring that clusters identified in one embedding are also captured in the other model. We provide numerical results demonstrating the SPEC's application to compare and align embeddings on large-scale datasets such as ImageNet and MS-COCO. The project page is available at https://mjalali.github.io/SPEC/. | [
{
"version": "v1",
"created": "Fri, 06 Jun 2025 16:50:37 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 12:12:39 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Jalali",
"Mohammad",
""
],
[
"Nia",
"Bahar Dibaei",
""
],
[
"Farnia",
"Farzan",
""
]
] |
2506.08141 | B Lekshmi | B Lekshmi, Sushanta Tripathy, Kiran Jain and Alexei Pevtsov | Dynamics of Subsurface Flows in Solar Active Regions During the May 2024 Storm | 12 pages, 6 figures, accepted for publication in The Astrophysical Journal | ApJ 987 134 (2025) | 10.3847/1538-4357/ade2e3 | null | astro-ph.SR | http://creativecommons.org/licenses/by/4.0/ | In May 2024, the Sun exhibited intense magnetic activity, marked by numerous high-intensity flares resulting from the interaction and merging of NOAA ARs 13664 and 13668 in the southern hemisphere and AR 13663 in the northern hemisphere. Notably, AR 13664 displayed an extended lifetime, remaining visible after a full solar rotation and continuing to produce significant flaring activity. In this study, we investigate the evolution of sub-photospheric plasma flows associated with these ARs during their disk passage using ring-diagram analysis of SDO/HMI Dopplergrams. We analyze flow divergence, vorticity, and kinetic helicity across depths from the surface to 25 Mm, revealing pronounced temporal and depth-dependent variations. Our observations indicate that the majority of flares occur on the days when the Normalized Helicity Gradient Variance, a measure of kinetic helicity spread, peaks or on the following day. Furthermore, we examine the relationship between subsurface flow dynamics and surface magnetic properties of these complex active regions to understand the interaction between them. | [
{
"version": "v1",
"created": "Mon, 09 Jun 2025 18:42:49 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 17:19:26 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Lekshmi",
"B",
""
],
[
"Tripathy",
"Sushanta",
""
],
[
"Jain",
"Kiran",
""
],
[
"Pevtsov",
"Alexei",
""
]
] |
2506.08320 | Aditya Patwardhan | Vivek Vaidya, Aditya Patwardhan, Ashish Kundu | How Good LLM-Generated Password Policies Are? | 11 pages, 2 Tables, 9 figures, 3 Algorithms | null | null | null | cs.CR cs.AI | http://creativecommons.org/licenses/by/4.0/ | Generative AI technologies, particularly Large Language Models (LLMs), are rapidly being adopted across industry, academia, and government sectors, owing to their remarkable capabilities in natural language processing. However, despite their strengths, the inconsistency and unpredictability of LLM outputs present substantial challenges, especially in security-critical domains such as access control. One critical issue that emerges prominently is the consistency of LLM-generated responses, which is paramount for ensuring secure and reliable operations.
In this paper, we study the application of LLMs within the context of Cybersecurity Access Control Systems. Specifically, we investigate the consistency and accuracy of LLM-generated password policies, translating natural language prompts into executable pwquality.conf configuration files. Our experimental methodology adopts two distinct approaches: firstly, we utilize pre-trained LLMs to generate configuration files purely from natural language prompts without additional guidance. Secondly, we provide these models with official pwquality.conf documentation to serve as an informative baseline. We systematically assess the soundness, accuracy, and consistency of these AI-generated configurations. Our findings underscore significant challenges in the current generation of LLMs and contribute valuable insights into refining the deployment of LLMs in Access Control Systems. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 01:12:31 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 23:34:36 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Vaidya",
"Vivek",
""
],
[
"Patwardhan",
"Aditya",
""
],
[
"Kundu",
"Ashish",
""
]
] |
2506.08713 | Fariz Ikhwantri | Fariz Ikhwantri, Dusica Marijan | Explainable Compliance Detection with Multi-Hop Natural Language Inference on Assurance Case Structure | null | null | null | null | cs.CL cs.SE | http://creativecommons.org/licenses/by/4.0/ | Ensuring complex systems meet regulations typically requires checking the validity of assurance cases through a claim-argument-evidence framework. Some challenges in this process include the complicated nature of legal and technical texts, the need for model explanations, and limited access to assurance case data. We propose a compliance detection approach based on Natural Language Inference (NLI): EXplainable CompLiance detection with Argumentative Inference of Multi-hop reasoning (EXCLAIM). We formulate the claim-argument-evidence structure of an assurance case as a multi-hop inference for explainable and traceable compliance detection. We address the limited number of assurance cases by generating them using large language models (LLMs). We introduce metrics that measure the coverage and structural consistency. We demonstrate the effectiveness of the generated assurance case from GDPR requirements in a multi-hop inference task as a case study. Our results highlight the potential of NLI-based approaches in automating the regulatory compliance process. | [
{
"version": "v1",
"created": "Tue, 10 Jun 2025 11:56:06 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 13:39:37 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Ikhwantri",
"Fariz",
""
],
[
"Marijan",
"Dusica",
""
]
] |
2506.09395 | Missagh Mehdipour | Missagh Mehdipour, Jelle S. Kaastra, Megan E. Eckart, Liyi Gu, Ralf Ballhausen, Ehud Behar, Camille M. Diez, Keigo Fukumura, Matteo Guainazzi, Kouichi Hagino, Timothy R. Kallman, Erin Kara, Chen Li, Jon M. Miller, Misaki Mizumoto, Hirofumi Noda, Shoji Ogawa, Christos Panagiotou, Atsushi Tanimoto, Keqin Zhao | Delving into the depths of NGC 3783 with XRISM. I. Kinematic and ionization structure of the highly ionized outflows | Accepted for publication in A&A, 10 pages, 6 figures | null | null | null | astro-ph.HE astro-ph.GA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present our study of the XRISM observation of the Seyfert-1 galaxy NGC 3783. XRISM's Resolve microcalorimeter has enabled, for the first time, a detailed characterization of the highly ionized outflows in this active galactic nucleus. Our analysis constrains their outflow and turbulent velocities, along with their ionization parameter ($\xi$) and column density ($N_{\rm H}$). The high-resolution Resolve spectrum reveals a distinct series of Fe absorption lines between 6.4 and 7.8 keV, ranging from Fe XVIII to Fe XXVI. At lower energies, absorption features from Si, S, and Ar are also detected. Our spectroscopy and photoionization modeling of the time-averaged Resolve spectrum uncovers six outflow components, five of which exhibit relatively narrow absorption lines with outflow velocities ranging from 560 to 1170 km/s. In addition, a broad absorption feature is detected, which is consistent with Fe XXVI outflowing at 14,300 km/s (0.05 $c$). The kinetic luminosity of this component is 0.8-3% of the bolometric luminosity. Our analysis of the Resolve spectrum shows that more highly ionized absorption lines are intrinsically broader than those of lower-ionization species, indicating that the turbulent velocity of the six outflow components (ranging from 0 to 3500 km/s) increases with $\xi$. Furthermore, we find that the $N_{\rm H}$ of the outflows generally declines with $\xi$ up to $\log \xi = 3.2$ but rises beyond this point, suggesting a complex ionization structure. The absorption profile of the Fe XXV resonance line is intriguingly similar to UV absorption lines (Ly$\alpha$ and C IV) observed by the HST, from which we infer that the outflows are clumpy in nature. Our XRISM/Resolve results support a "hybrid wind" scenario in which the observed outflows have multiple origins and driving mechanisms. We explore various interpretations of our findings within AGN wind models. | [
{
"version": "v1",
"created": "Wed, 11 Jun 2025 04:54:57 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 01:02:42 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Mehdipour",
"Missagh",
""
],
[
"Kaastra",
"Jelle S.",
""
],
[
"Eckart",
"Megan E.",
""
],
[
"Gu",
"Liyi",
""
],
[
"Ballhausen",
"Ralf",
""
],
[
"Behar",
"Ehud",
""
],
[
"Diez",
"Camille M.",
""
],
[
"Fukumura",
"Keigo",
""
],
[
"Guainazzi",
"Matteo",
""
],
[
"Hagino",
"Kouichi",
""
],
[
"Kallman",
"Timothy R.",
""
],
[
"Kara",
"Erin",
""
],
[
"Li",
"Chen",
""
],
[
"Miller",
"Jon M.",
""
],
[
"Mizumoto",
"Misaki",
""
],
[
"Noda",
"Hirofumi",
""
],
[
"Ogawa",
"Shoji",
""
],
[
"Panagiotou",
"Christos",
""
],
[
"Tanimoto",
"Atsushi",
""
],
[
"Zhao",
"Keqin",
""
]
] |
2506.09524 | Xueyuan Wan | Inkang Kim and Xueyuan Wan | Positivity of simplicial volume for closed nonpositively curved four-manifolds with nonzero Euler characteristic | 12 pages. Added further details. Comments are welcome | null | null | MPIM-Bonn-2025 | math.GT math.DG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, by employing the Gauss-Bonnet theorem for Riemannian simplices due to Allendoerfer and Weil, we show that if a closed nonpositively curved $4$-manifold has nonzero Euler characteristic, then its simplicial volume is necessarily positive. This result partially resolves conjectures posed by Connell-Ruan-Wang and Gromov concerning the relationship between the simplicial volume and the Euler characteristic for four-dimensional manifolds. As an application, we show that if a closed nonpositively curved $4$-manifold has negative Ricci curvature, then its simplicial volume is positive, thereby confirming in dimension four another conjecture of Gromov on the positivity of simplicial volume. | [
{
"version": "v1",
"created": "Wed, 11 Jun 2025 08:49:47 GMT"
},
{
"version": "v2",
"created": "Mon, 16 Jun 2025 12:14:57 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Kim",
"Inkang",
""
],
[
"Wan",
"Xueyuan",
""
]
] |
2506.09612 | Mingxiao Li | Mingxiao Li, Mang Ning, Marie-Francine Moens | Consistent Story Generation with Asymmetry Zigzag Sampling | 17 pages, 9. figures | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Text-to-image generation models have made significant progress in producing high-quality images from textual descriptions, yet they continue to struggle with maintaining subject consistency across multiple images, a fundamental requirement for visual storytelling. Existing methods attempt to address this by either fine-tuning models on large-scale story visualization datasets, which is resource-intensive, or by using training-free techniques that share information across generations, which still yield limited success. In this paper, we introduce a novel training-free sampling strategy called Zigzag Sampling with Asymmetric Prompts and Visual Sharing to enhance subject consistency in visual story generation. Our approach proposes a zigzag sampling mechanism that alternates between asymmetric prompting to retain subject characteristics, while a visual sharing module transfers visual cues across generated images to %further enforce consistency. Experimental results, based on both quantitative metrics and qualitative evaluations, demonstrate that our method significantly outperforms previous approaches in generating coherent and consistent visual stories. The code is available at https://github.com/Mingxiao-Li/Asymmetry-Zigzag-StoryDiffusion. | [
{
"version": "v1",
"created": "Wed, 11 Jun 2025 11:14:27 GMT"
},
{
"version": "v2",
"created": "Thu, 12 Jun 2025 13:02:01 GMT"
},
{
"version": "v3",
"created": "Wed, 02 Jul 2025 20:03:42 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Li",
"Mingxiao",
""
],
[
"Ning",
"Mang",
""
],
[
"Moens",
"Marie-Francine",
""
]
] |
2506.09993 | Jaewon Min | Jaewon Min, Jin Hyeon Kim, Paul Hyunbin Cho, Jaeeun Lee, Jihye Park, Minkyu Park, Sangpil Kim, Hyunhee Park, Seungryong Kim | Text-Aware Image Restoration with Diffusion Models | Project page: https://cvlab-kaist.github.io/TAIR/ | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Image restoration aims to recover degraded images. However, existing diffusion-based restoration methods, despite great success in natural image restoration, often struggle to faithfully reconstruct textual regions in degraded images. Those methods frequently generate plausible but incorrect text-like patterns, a phenomenon we refer to as text-image hallucination. In this paper, we introduce Text-Aware Image Restoration (TAIR), a novel restoration task that requires the simultaneous recovery of visual contents and textual fidelity. To tackle this task, we present SA-Text, a large-scale benchmark of 100K high-quality scene images densely annotated with diverse and complex text instances. Furthermore, we propose a multi-task diffusion framework, called TeReDiff, that integrates internal features from diffusion models into a text-spotting module, enabling both components to benefit from joint training. This allows for the extraction of rich text representations, which are utilized as prompts in subsequent denoising steps. Extensive experiments demonstrate that our approach consistently outperforms state-of-the-art restoration methods, achieving significant gains in text recognition accuracy. See our project page: https://cvlab-kaist.github.io/TAIR/ | [
{
"version": "v1",
"created": "Wed, 11 Jun 2025 17:59:46 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 06:39:43 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Min",
"Jaewon",
""
],
[
"Kim",
"Jin Hyeon",
""
],
[
"Cho",
"Paul Hyunbin",
""
],
[
"Lee",
"Jaeeun",
""
],
[
"Park",
"Jihye",
""
],
[
"Park",
"Minkyu",
""
],
[
"Kim",
"Sangpil",
""
],
[
"Park",
"Hyunhee",
""
],
[
"Kim",
"Seungryong",
""
]
] |
2506.10374 | Daniel McMorrow | Daniel McMorrow and Jonathan Scarlett | Optimal Non-Adaptive Group Testing with One-Sided Error Guarantees | null | null | null | null | cs.IT math.IT math.ST stat.TH | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The group testing problem consists of determining a sparse subset of defective items from within a larger set of items via a series of tests, where each test outcome indicates whether at least one defective item is included in the test. We study the approximate recovery setting, where the recovery criterion of the defective set is relaxed to allow a small number of items to be misclassified. In particular, we consider one-sided approximate recovery criteria, where we allow either only false negative or only false positive misclassifications. Under false negatives only (i.e., finding a subset of defectives), we show that there exists an algorithm matching the optimal threshold of two-sided approximate recovery. Under false positives only (i.e., finding a superset of the defectives), we provide a converse bound showing that the better of two existing algorithms is optimal. | [
{
"version": "v1",
"created": "Thu, 12 Jun 2025 05:53:53 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 02:13:31 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"McMorrow",
"Daniel",
""
],
[
"Scarlett",
"Jonathan",
""
]
] |
2506.10471 | Naoki Matsumoto | Kengo Enami, Naoki Matsumoto, Takamasa Yashima | Contributions to conjectures in planar graphs: Induced Subgraphs, Treewidth, and Dominating Sets | null | null | null | null | cs.DM math.CO | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Two of the most prominent unresolved conjectures in graph theory, the Albertson-Berman conjecture and the Matheson-Tarjan conjecture, have been extensively studied by many researchers.
(AB) Every planar graph of order $n$ has an induced forest of order at least $\frac{n}{2}$.
(MT) Every plane triangulation of sufficiently large order $n$ has a dominating set of cardinality at most $\frac{n}{4}$.
Although partial progress and weaker bounds are known, both conjectures remain unsolved. To shed further light on them, researchers have explored a variety of related notions and generalizations. In this paper, we clarify relations among several of these notions, most notably connected domination and induced outerplanar subgraphs, and investigate the corresponding open problems. Furthermore, we construct an infinite family of plane triangulations of order $n$ whose connected domination number exceeds $n/3$. This construction gives a negative answer to a question of Bradshaw et al. [SIAM J. Discrete Math. 36 (2022) 1416-1435], who asked whether the maxleaf number of every plane triangulation of order $n$ is at least $2n/3$. We also obtain new results on induced subgraphs with bounded treewidth and induced outerplanar subgraphs. | [
{
"version": "v1",
"created": "Thu, 12 Jun 2025 08:22:57 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 14:30:26 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Enami",
"Kengo",
""
],
[
"Matsumoto",
"Naoki",
""
],
[
"Yashima",
"Takamasa",
""
]
] |
2506.10799 | Wuji Wang | Wuji Wang, Carlos De Breuck, Dominika Wylezalek, Matthew D. Lehnert, Andreas L. Faisst, Andrey Vayner, Nicole Nesvadba, Jo\"el Vernet, Pranav Kukreti, Daniel Stern | Gas-Poor Hosts and Gas-Rich Companions of $z\approx$3.5 Radio AGN: ALMA insights into jet triggering and feedback | Accepted for publication in ApJL; 10 pages, 4 figures, and 1 table in main text; update author list | null | null | null | astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | Cold gaseous systems play important roles in galaxy evolution by possibly providing fuel to ignite active galactic nuclei (AGN) activity and star-formation. In this work, we analyze [CII]$158\rm \mu m$ and continuum observations from ALMA for a sample of four radio AGN at $z \approx 3.5$, focusing on eight associated companion cloud systems discovered within projected distances of tens of kiloparsecs or less. The spatial distribution of these companions indicates that the majority of cold gas is not located at the AGN position, i.e., not in their host galaxies. With the assistance of [CII] at 0.2" resolution, we further confirm the gas-poor nature of the hosts by re-analyzing archival [CI] (a tracer of H$_{2}$) at $\sim2"$ resolution. Our sample has [CII] luminosities in a range of $2.8\times10^{8}<L_{\rm [CII]}/L_{\odot}<4.2\times10^{9}$. The $L_{\rm [CII]}/L_{\rm IR}$ ratio, $\sim 9.4\times10^{-4}$, is consistent with sources discussed in the literature. Our findings show the gas-poor radio AGN hosts have nearby gas-rich companions. We propose that these companions, which may be stripped clouds in merger process and by interaction with the energy output from the AGN, could be a trigger of radio-loud AGN. They may also be a signature of negative AGN feedback (e.g., shock heating) on these infalling companions and on the host galaxy. In general, our analysis shows that powerful AGN at and before Cosmic Noon are impacting and being impacted by cold gaseous clouds in their circum-galactic or proto-intracluster media. | [
{
"version": "v1",
"created": "Thu, 12 Jun 2025 15:14:14 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 21:36:06 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Wang",
"Wuji",
""
],
[
"De Breuck",
"Carlos",
""
],
[
"Wylezalek",
"Dominika",
""
],
[
"Lehnert",
"Matthew D.",
""
],
[
"Faisst",
"Andreas L.",
""
],
[
"Vayner",
"Andrey",
""
],
[
"Nesvadba",
"Nicole",
""
],
[
"Vernet",
"Joël",
""
],
[
"Kukreti",
"Pranav",
""
],
[
"Stern",
"Daniel",
""
]
] |
2506.10931 | Konstantina Koliogeorgi Koliogeorgi | Melina Soysal, Konstantina Koliogeorgi, Can Firtina, Nika Mansouri Ghiasi, Rakesh Nadig, Haiyu Mao, Geraldo F. Oliveira, Yu Liang, Klea Zambaku, Mohammad Sadrosadati, Onur Mutlu | MARS: Processing-In-Memory Acceleration of Raw Signal Genome Analysis Inside the Storage Subsystem | null | null | 10.1145/3721145.3730428 | null | cs.AR q-bio.GN | http://creativecommons.org/licenses/by/4.0/ | Raw signal genome analysis (RSGA) has emerged as a promising approach to enable real-time genome analysis by directly analyzing raw electrical signals. However, rapid advancements in sequencing technologies make it increasingly difficult for software-based RSGA to match the throughput of raw signal generation. This paper demonstrates that while hardware acceleration techniques can significantly accelerate RSGA, the high volume of genomic data shifts the performance and energy bottleneck from computation to I/O data movement. As sequencing throughput increases, I/O overhead becomes the main contributor to both runtime and energy consumption. Therefore, there is a need to design a high-performance, energy-efficient system for RSGA that can both alleviate the data movement bottleneck and provide large acceleration capabilities. We propose MARS, a storage-centric system that leverages the heterogeneous resources within modern storage systems (e.g., storage-internal DRAM, storage controller, flash chips) alongside their large storage capacity to tackle both data movement and computational overheads of RSGA in an area-efficient and low-cost manner. MARS accelerates RSGA through a novel hardware/software co-design approach. First, MARS modifies the RSGA pipeline via two filtering mechanisms and a quantization scheme, reducing hardware demands and optimizing for in-storage execution. Second, MARS accelerates the RSGA steps directly within the storage by leveraging both Processing-Near-Memory and Processing-Using-Memory paradigms. Third, MARS orchestrates the execution of all steps to fully exploit in-storage parallelism and minimize data movement. Our evaluation shows that MARS outperforms basecalling-based software and hardware-accelerated state-of-the-art read mapping pipelines by 93x and 40x, on average across different datasets, while reducing their energy consumption by 427x and 72x. | [
{
"version": "v1",
"created": "Thu, 12 Jun 2025 17:38:12 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 13:48:25 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Soysal",
"Melina",
""
],
[
"Koliogeorgi",
"Konstantina",
""
],
[
"Firtina",
"Can",
""
],
[
"Ghiasi",
"Nika Mansouri",
""
],
[
"Nadig",
"Rakesh",
""
],
[
"Mao",
"Haiyu",
""
],
[
"Oliveira",
"Geraldo F.",
""
],
[
"Liang",
"Yu",
""
],
[
"Zambaku",
"Klea",
""
],
[
"Sadrosadati",
"Mohammad",
""
],
[
"Mutlu",
"Onur",
""
]
] |
2506.11369 | Shuhao Jiao | Shuhao Jiao, Hernando Ombao and Ian W. McKeague | Filtrated Grouping in Multiple Functional Regression | null | null | null | null | stat.ME stat.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | To understand and communicate the risk of chronic joint disease associated with aging, it is essential to investigate how age is associated with gait patterns, particularly through gait angular kinematics. Motivated by this need, and by the critical role of joint coordination in gait, we propose a novel covariate grouping framework within the context of multiple functional regression, where a scalar response is linked to multiple functional covariates. We apply this approach to investigate the relationship between chronological age and gait angular kinematics, aiming to uncover biomechanical patterns that signal age-related gait pattern evolution. Specifically, we develop a forest-structured covariate grouping framework in which different functional covariates are aggregated hierarchically based on the level of coefficient homogeneity. This approach allows for the analysis of both common and idiosyncratic effects of covariates in a nuanced, multi-resolution manner. The identification of the forest structure is entirely data-driven and requires no prior knowledge, providing valuable insights into the interdependence among covariates. Compared to existing methods, the proposed regression framework demonstrates superior predictive power and offers more insightful interpretability on joint coordination. In addition, the proposed framework is broadly applicable and can be readily extended to analyze multivariate functional data in other scientific domains. | [
{
"version": "v1",
"created": "Fri, 13 Jun 2025 00:08:41 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 14:44:44 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Jiao",
"Shuhao",
""
],
[
"Ombao",
"Hernando",
""
],
[
"McKeague",
"Ian W.",
""
]
] |
2506.12479 | Jiawei Shao | Hongjun An, Wenhan Hu, Sida Huang, Siqi Huang, Ruanjun Li, Yuanzhi Liang, Jiawei Shao, Yiliang Song, Zihan Wang, Cheng Yuan, Chi Zhang, Hongyuan Zhang, Wenhao Zhuang, Xuelong Li | AI Flow: Perspectives, Scenarios, and Approaches | Authors are with Institute of Artificial Intelligence (TeleAI), China Telecom, China. Author names are listed alphabetically by surname. This work was conducted at TeleAI, facilitated by Dr. Jiawei Shao (e-mail: [email protected]) under the leadership of Prof. Xuelong Li. The corresponding author is Prof. Xuelong Li (e-mail: xuelong [email protected]), the CTO and Chief Scientist of China Telecom | null | null | null | cs.AI cs.CL cs.CV cs.DC eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Pioneered by the foundational information theory by Claude Shannon and the visionary framework of machine intelligence by Alan Turing, the convergent evolution of information and communication technologies (IT/CT) has created an unbroken wave of connectivity and computation. This synergy has sparked a technological revolution, now reaching its peak with large artificial intelligence (AI) models that are reshaping industries and redefining human-machine collaboration. However, the realization of ubiquitous intelligence faces considerable challenges due to substantial resource consumption in large models and high communication bandwidth demands. To address these challenges, AI Flow has been introduced as a multidisciplinary framework that integrates cutting-edge IT and CT advancements, with a particular emphasis on the following three key points. First, device-edge-cloud framework serves as the foundation, which integrates end devices, edge servers, and cloud clusters to optimize scalability and efficiency for low-latency model inference. Second, we introduce the concept of familial models, which refers to a series of different-sized models with aligned hidden features, enabling effective collaboration and the flexibility to adapt to varying resource constraints and dynamic scenarios. Third, connectivity- and interaction-based intelligence emergence is a novel paradigm of AI Flow. By leveraging communication networks to enhance connectivity, the collaboration among AI models across heterogeneous nodes achieves emergent intelligence that surpasses the capability of any single model. The innovations of AI Flow provide enhanced intelligence, timely responsiveness, and ubiquitous accessibility to AI services, paving the way for the tighter fusion of AI techniques and communication systems. | [
{
"version": "v1",
"created": "Sat, 14 Jun 2025 12:43:07 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 12:11:34 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"An",
"Hongjun",
""
],
[
"Hu",
"Wenhan",
""
],
[
"Huang",
"Sida",
""
],
[
"Huang",
"Siqi",
""
],
[
"Li",
"Ruanjun",
""
],
[
"Liang",
"Yuanzhi",
""
],
[
"Shao",
"Jiawei",
""
],
[
"Song",
"Yiliang",
""
],
[
"Wang",
"Zihan",
""
],
[
"Yuan",
"Cheng",
""
],
[
"Zhang",
"Chi",
""
],
[
"Zhang",
"Hongyuan",
""
],
[
"Zhuang",
"Wenhao",
""
],
[
"Li",
"Xuelong",
""
]
] |
2506.12535 | Susovan Pramanik | Susovan Pramanik (Harish-Chandra Research Institute, India) | Anisotropic Calder\'{o}n problem of a nearly Laplace-Beltrami operator of order $2+$ | 18 pages | null | null | null | math.AP | http://creativecommons.org/licenses/by/4.0/ | This paper investigates the anisotropic Calder\'{o}n problem for Logarithemic Laplacian, on closed Riemannian manifolds, which could be considered as near Laplace operator. We demonstrate that the Cauchy data set recovers the geometry of a closed Riemannian manifold up to standard gauge. | [
{
"version": "v1",
"created": "Sat, 14 Jun 2025 15:15:37 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 04:22:57 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Pramanik",
"Susovan",
"",
"Harish-Chandra Research Institute, India"
]
] |
2506.12999 | Brody Lynch | Brody Lynch | Equidistribution of realizable Steinitz classes for cyclic Kummer extensions | 24 pages, 1 figure. V2 has minor typo fixed in the introduction and a simpler proof of Lemma 6.1 | null | null | null | math.NT | http://creativecommons.org/licenses/by/4.0/ | Let $\ell$ be prime, and $K$ be a number field containing the $\ell$-th roots of unity. We use techniques from classical algebraic number theory to prove that the Steinitz classes of $\mathbb Z/\ell\mathbb Z$ extensions of $K$ are equidistributed among realizable classes in the ideal class group of $K$. For $\ell = 2$, this was proved by Kable and Wright using the deep theory of prehomogeneous vector spaces. Foster proved that Steinitz classes are uniformly distributed between realizable classes for tamely ramified elementary-$m$ extensions using the theory of Galois modules; our approach eliminates this tameness hypothesis. | [
{
"version": "v1",
"created": "Sun, 15 Jun 2025 23:54:59 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 17:19:33 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Lynch",
"Brody",
""
]
] |
2506.13359 | Markus Frankenbach | Markus Frankenbach, Marc Ritter, Mathias Pelz, Nepomuk Ritz, Jan von Delft and Anxiang Ge | Compressing local vertex functions from the multipoint numerical renormalization group using quantics tensor cross interpolation | 19 pages, 12 figures, 5 tables | null | null | null | cond-mat.str-el | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The multipoint numerical renormalization group (mpNRG) is a powerful impurity solver that provides accurate spectral data useful for computing local, dynamic correlation functions in imaginary or real frequencies non-perturbatively across a wide range of interactions and temperatures. It gives access to a local, non-perturbative four-point vertex in imaginary and real frequencies, which can be used as input for subsequent computations such as diagrammatic extensions of dynamical mean--field theory. However, computing and manipulating the real-frequency four-point vertex on large, dense grids quickly becomes numerically challenging when the density and/or the extent of the frequency grid is increased. In this paper, we compute four-point vertices in a strongly compressed quantics tensor train format using quantics tensor cross interpolation, starting from discrete partial spectral functions obtained from mpNRG. This enables evaluations of the vertex on frequency grids with resolutions far beyond the reach of previous implementations. We benchmark this approach on the four-point vertex of the single-impurity Anderson model across a wide range of physical parameters, both in its full form and its asymptotic decomposition. For imaginary frequencies, the full vertex can be represented to an accuracy on the order of $2\cdot 10^{-3}$ with maximum bond dimensions not exceeding 120. The more complex full real-frequency vertex requires maximum bond dimensions not exceeding 170 for an accuracy of $\lesssim 2\%$. Our work marks another step toward tensor-train-based diagrammatic calculations for correlated electronic lattice models starting from a local, non-perturbative mpNRG vertex. | [
{
"version": "v1",
"created": "Mon, 16 Jun 2025 11:00:05 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 18:54:58 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Frankenbach",
"Markus",
""
],
[
"Ritter",
"Marc",
""
],
[
"Pelz",
"Mathias",
""
],
[
"Ritz",
"Nepomuk",
""
],
[
"von Delft",
"Jan",
""
],
[
"Ge",
"Anxiang",
""
]
] |
2506.13522 | Philip Massey | Philip Massey, Sarah Bodansky, Laura R. Penny, and Kathryn F. Neugent | A New Determination of the Mass of NGC 3603-A1: the Most Massive Binary Known? | Accepted by the Astrophysical Journal. This revision includes a previously overlooked paper on WR21a in our discussion | null | null | null | astro-ph.SR astro-ph.GA | http://creativecommons.org/licenses/by-nc-sa/4.0/ | The star NGC 3603-A1 has long been known to be a very massive binary, consisting of a pair of O2-3If*/WN5-6 stars, which show Wolf-Rayet-like emission due to their luminosities being near the Eddington limit. The system has been poorly characterized until now, due to the difficulties of obtaining reliable radial velocities from broad, blended emission lines and the extreme crowding in the cluster. However, previously unpublished archival HST/STIS spectra revealed that some of the upper Balmer lines (seen in absorption) are well separated at favorable orbital phases, prompting us to obtain our own carefully-timed new HST/STIS spectra, which we have analyzed along with the older data. Radial velocities measured from these spectra allow us to obtain an orbit for this 3.77298 day system. We also used archival STIS imaging of the cluster to obtain a more accurate light curve for this eclipsing system, which we then modeled, yielding the orbital inclination and providing values for the stellar radii and temperatures. Together, these data show that the NGC 3603-A1 system consists of a 93.3+/-11.0 Mo O3If*/WN6 primary with an effective temperature of 37,000 K, and a 70.4+/-9.3 Mo O3If*/WN5 secondary that is slightly hotter, 42,000 K. Although a more massive binary is known in the LMC, NGC 3603-A1 is as massive as any binary known in our own Galaxy for which a direct measurement of its mass has been made. The secondary has been spun up by mass accretion from the primary, and we discuss the evolutionary status of this intriguing system. | [
{
"version": "v1",
"created": "Mon, 16 Jun 2025 14:15:14 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 15:17:34 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Massey",
"Philip",
""
],
[
"Bodansky",
"Sarah",
""
],
[
"Penny",
"Laura R.",
""
],
[
"Neugent",
"Kathryn F.",
""
]
] |
2506.13885 | Chao Li | Chao Li and Boyu Zhang | Covering instability for the existence of positive scalar curvature metrics | Minor edits, more references added | null | null | null | math.DG math.GT math.MG | http://creativecommons.org/licenses/by/4.0/ | We show that a closed non-orientable $3$-manifold admits a positive scalar curvature metric if and only if its orientation double cover does; however, for each $4\le n\le 7$, there exist infinitely many smooth non-orientable $n$-manifolds $M$ that are mutually non-homotopy equivalent, such that the orientation double cover of $M$ admits positive scalar curvature metrics, but every closed smooth manifold that is homotopy equivalent to $M$ cannot admit positive scalar curvature metrics. These examples were first introduced by Alpert-Balitskiy-Guth in the study of Urysohn widths.
To prove the nonexistence result, we extend the Schoen-Yau inductive descent approach to non-orientable manifolds. We also discuss band width estimates and the notion of enlargeability for non-orientable PSC manifolds. | [
{
"version": "v1",
"created": "Mon, 16 Jun 2025 18:05:31 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 18:48:05 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Li",
"Chao",
""
],
[
"Zhang",
"Boyu",
""
]
] |
2506.13972 | Zhiqi Wang | Zhiqi Wang, Chengyu Zhang, Yuetian Chen, Nathalie Baracaldo, Swanand Kadhe, Lei Yu | Membership Inference Attacks as Privacy Tools: Reliability, Disparity and Ensemble | 22 pages, To appear at ACM CCS 2025 | null | 10.1145/3719027.3744818 | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Membership inference attacks (MIAs) pose a significant threat to the privacy of machine learning models and are widely used as tools for privacy assessment, auditing, and machine unlearning. While prior MIA research has primarily focused on performance metrics such as AUC, accuracy, and TPR@low FPR - either by developing new methods to enhance these metrics or using them to evaluate privacy solutions - we found that it overlooks the disparities among different attacks. These disparities, both between distinct attack methods and between multiple instantiations of the same method, have crucial implications for the reliability and completeness of MIAs as privacy evaluation tools. In this paper, we systematically investigate these disparities through a novel framework based on coverage and stability analysis. Extensive experiments reveal significant disparities in MIAs, their potential causes, and their broader implications for privacy evaluation. To address these challenges, we propose an ensemble framework with three distinct strategies to harness the strengths of state-of-the-art MIAs while accounting for their disparities. This framework not only enables the construction of more powerful attacks but also provides a more robust and comprehensive methodology for privacy evaluation. | [
{
"version": "v1",
"created": "Mon, 16 Jun 2025 20:22:07 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 17:45:38 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Wang",
"Zhiqi",
""
],
[
"Zhang",
"Chengyu",
""
],
[
"Chen",
"Yuetian",
""
],
[
"Baracaldo",
"Nathalie",
""
],
[
"Kadhe",
"Swanand",
""
],
[
"Yu",
"Lei",
""
]
] |
2506.14634 | Leah Von Der Heyde | Leah von der Heyde, Anna-Carolina Haensch, Bernd Wei{\ss}, Jessica Daikeler | AIn't Nothing But a Survey? Using Large Language Models for Coding German Open-Ended Survey Responses on Survey Motivation | to appear in Survey Research Methods | null | null | null | cs.CL cs.AI cs.CY | http://creativecommons.org/licenses/by/4.0/ | The recent development and wider accessibility of LLMs have spurred discussions about how they can be used in survey research, including classifying open-ended survey responses. Due to their linguistic capacities, it is possible that LLMs are an efficient alternative to time-consuming manual coding and the pre-training of supervised machine learning models. As most existing research on this topic has focused on English-language responses relating to non-complex topics or on single LLMs, it is unclear whether its findings generalize and how the quality of these classifications compares to established methods. In this study, we investigate to what extent different LLMs can be used to code open-ended survey responses in other contexts, using German data on reasons for survey participation as an example. We compare several state-of-the-art LLMs and several prompting approaches, and evaluate the LLMs' performance by using human expert codings. Overall performance differs greatly between LLMs, and only a fine-tuned LLM achieves satisfactory levels of predictive performance. Performance differences between prompting approaches are conditional on the LLM used. Finally, LLMs' unequal classification performance across different categories of reasons for survey participation results in different categorical distributions when not using fine-tuning. We discuss the implications of these findings, both for methodological research on coding open-ended responses and for their substantive analysis, and for practitioners processing or substantively analyzing such data. Finally, we highlight the many trade-offs researchers need to consider when choosing automated methods for open-ended response classification in the age of LLMs. In doing so, our study contributes to the growing body of research about the conditions under which LLMs can be efficiently, accurately, and reliably leveraged in survey research. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 15:28:53 GMT"
},
{
"version": "v2",
"created": "Wed, 18 Jun 2025 09:56:49 GMT"
},
{
"version": "v3",
"created": "Thu, 03 Jul 2025 07:58:18 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"von der Heyde",
"Leah",
""
],
[
"Haensch",
"Anna-Carolina",
""
],
[
"Weiß",
"Bernd",
""
],
[
"Daikeler",
"Jessica",
""
]
] |
2506.15427 | Yang He | Yang He, Artan Sheshmani | Geography of Landau-Ginzburg models and threefold syzygies | 61 pages, comments welcomed | null | null | null | math.AG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study the behavior of toric Landau-Ginzburg models under extremal contraction and minimal model program. We also establish a relation between the moduli space of toric Landau-Ginzburg models and the geography of central models. We conjecture that there is a correspondence between extremal contractions and minimal model program on Fano varieties, and degenerations of their associated toric Landau-Ginzburg models written explicitly. We prove the conjectures for smooth toric varieties, as well as general smooth Fano varieties in dimensions 2 and 3. As an application, we compute the elementary syzygies for smooth Fano threefolds. | [
{
"version": "v1",
"created": "Wed, 18 Jun 2025 13:02:22 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 14:15:43 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"He",
"Yang",
""
],
[
"Sheshmani",
"Artan",
""
]
] |
2506.15463 | Shweta Pal | Shweta Pal, Arun Kumar, Monika Agrawal | Effect of Signal Quantization on Performance Measures of a 1st Order One Dimensional Differential Microphone Array | 5 Pages with 6 figures and 1 table | null | null | null | eess.SP | http://creativecommons.org/licenses/by-nc-nd/4.0/ | In practical systems, recorded analog signals must be digitized for processing, introducing quantization as a critical aspect of data acquisition. While prior studies have examined quantization effects in various signal processing contexts, its impact on differential microphone arrays (DMAs), particularly in one-dimensional (1D) first-order configurations, remains unexplored. This paper investigates the influence of signal quantization on the performance of first-order 1D-DMAs across various beampatterns. An analytical expression for quantized beamformed output for a first-order 1D-DMA has been formulated. The effect of signal quantization has been studied on array performance measures such as the Beampattern, Directivity Factor (DF), Front-to-Back Ratio (FBR), and suppression depth at null points (SDN). Simulation results reveal that the beampattern shape remains structurally invariant across quantization bit depths, with quantization primarily affecting SDN. DF and FBR remain constant with the varying number of quantization bits. Additionally, SDN is shown to be frequency-independent, however, it increases with increasing quantization bit depths, enhancing interference suppression. The study also examines the effect of steering nulls across the azimuthal range, showing that SDN degrades as the null moves closer to the source look direction (0 degree), indicating reduced interference suppression. | [
{
"version": "v1",
"created": "Wed, 18 Jun 2025 13:55:01 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 11:18:46 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Pal",
"Shweta",
""
],
[
"Kumar",
"Arun",
""
],
[
"Agrawal",
"Monika",
""
]
] |
2506.15778 | Carlos Nunez | Niall Macpherson, Paul Merrikin, Carlos Nunez and Ricardo Stuardo | Twisted-Circle Compactifications of SQCD-like Theories and Holography | 21 pages, two figures | null | null | null | hep-th | http://creativecommons.org/licenses/by/4.0/ | We construct and analyse holographic duals to a class of four-dimensional N = 1 SU($N_c$) SQCD-like theories compactified on a circle with an R-symmetry twist. The setup originates from type IIB backgrounds previously proposed as duals to SQCD with $N_f$ fundamental flavours. The U(1) R-symmetry is anomaly-free only if $ N_f = 2N_c$. We implement a supersymmetric twisted-circle reduction, holographically realised through a smoothly shrinking $S^1$ fibered over the internal U(1)$_R$ direction. We obtain new regular type IIB supergravity backgrounds that are valid only if the condition $N_f = 2N_c$ is satisfied--mirroring the anomaly cancellation requirement in the field theory. We compute various field-theoretic observables--including the Chern-Simons level, Wilson loop and the holographic central charge--showing the emergence of a 3D ${\cal N} = 2$ gapped phase consistent with a Chern-Simons TQFT. This work highlights the interplay between anomalies, supersymmetry, and geometry in the holographic realisation of compactified gauge theories with fundamental matter. | [
{
"version": "v1",
"created": "Wed, 18 Jun 2025 18:00:03 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 10:18:25 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Macpherson",
"Niall",
""
],
[
"Merrikin",
"Paul",
""
],
[
"Nunez",
"Carlos",
""
],
[
"Stuardo",
"Ricardo",
""
]
] |
2506.15830 | Riccardo Di Sipio | Riccardo Di Sipio | Rethinking LLM Training through Information Geometry and Quantum Metrics | 9 pages, 1 figure(s) | null | null | null | cs.CL quant-ph | http://creativecommons.org/licenses/by/4.0/ | Optimization in large language models (LLMs) unfolds over high-dimensional parameter spaces with non-Euclidean structure. Information geometry frames this landscape using the Fisher information metric, enabling more principled learning via natural gradient descent. Though often impractical, this geometric lens clarifies phenomena such as sharp minima, generalization, and observed scaling laws. We argue that curvature-aware approaches deepen our understanding of LLM training. Finally, we speculate on quantum analogies based on the Fubini-Study metric and Quantum Fisher Information, hinting at efficient optimization in quantum-enhanced systems. | [
{
"version": "v1",
"created": "Wed, 18 Jun 2025 19:17:47 GMT"
},
{
"version": "v2",
"created": "Thu, 26 Jun 2025 16:14:42 GMT"
},
{
"version": "v3",
"created": "Wed, 02 Jul 2025 21:36:55 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Di Sipio",
"Riccardo",
""
]
] |
2506.15854 | Abdolazim Rezaei | Abdolazim Rezaei, Mehdi Sookhak, Ahmad Patooghy | Privacy-Preserving in Connected and Autonomous Vehicles Through Vision to Text Transformation | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Connected and Autonomous Vehicles (CAVs) rely on a range of devices that often process privacy-sensitive data. Among these, roadside units play a critical role particularly through the use of AI-equipped (AIE) cameras for applications such as violation detection. However, the privacy risks associated with captured imagery remain a major concern, as such data can be misused for identity theft, profiling, or unauthorized commercial purposes. While traditional techniques such as face blurring and obfuscation have been applied to mitigate privacy risks, individual privacy remains at risk, as individuals can still be tracked using other features such as their clothing. This paper introduces a novel privacy-preserving framework that leverages feedback-based reinforcement learning (RL) and vision-language models (VLMs) to protect sensitive visual information captured by AIE cameras. The main idea is to convert images into semantically equivalent textual descriptions, ensuring that scene-relevant information is retained while visual privacy is preserved. A hierarchical RL strategy is employed to iteratively refine the generated text, enhancing both semantic accuracy and privacy. Evaluation results demonstrate significant improvements in both privacy protection and textual quality, with the Unique Word Count increasing by approximately 77\% and Detail Density by around 50\% compared to existing approaches. | [
{
"version": "v1",
"created": "Wed, 18 Jun 2025 20:02:24 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 18:31:53 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Rezaei",
"Abdolazim",
""
],
[
"Sookhak",
"Mehdi",
""
],
[
"Patooghy",
"Ahmad",
""
]
] |
2506.16676 | Mark Adams | Mark F. Adams, Jin Chen, Benjamin Sturdevant | Fast solvers for Tokamak fluid models with PETSC -- Part I | null | null | null | null | physics.plasm-ph cs.PF | http://creativecommons.org/publicdomain/zero/1.0/ | This work begins the development of fast, scalable solvers for scientific and engineering-relevant magnetohydrodynamics (MHD) models of tokamaks using multigrid methods. These tokamak models are characterized by a distinguished direction in the toroidal coordinate that is partially aligned with the magnetic guide field, which dominates the plasma dynamics. All tokamak models exploit this structure, for example, NIMROD at https://nimrodteam.org uses $2D$, unstructured, high-order finite elements in the poloidal plane with Fourier modes in the toroidal coordinate, and the $3D$, extended MHD code \textit{M3D-C1}\footnote{https://m3dc1.pppl.gov} uses $2D$, unstructured $C^1$ elements in the poloidal plane with cubic Hermite functions in the toroidal direction. This structure suggests addressing the toroidal coordinate first, which \textit{NIMROD} does at the formulation level, but the \textit{M3D-C1} approach leaves in the algebraic system to be solved at each time step in an implicit time integrator. This work addressed the toroidal coordinate in the \textit{M3D-C1} velocity solve by adding semi-coarsening multigrid to the existing PETSC at https://petsc.org -- Portable, Extensible Toolkit for Scientific Computation -- block Jacobi solver, with the addition of little new code that allows for smaller Jacobi subdomains that are better suited to contemporary, highly parallel, hardware. Competitive performance of this new solver configuration is demonstrated on a self-consistent runaway electron model of a SPARC at https://cfs.energy/technology/sparc disruption, and the next steps in the development of this new approach are outlined. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 01:36:08 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 19:08:22 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Adams",
"Mark F.",
""
],
[
"Chen",
"Jin",
""
],
[
"Sturdevant",
"Benjamin",
""
]
] |
2506.16715 | Markus Abel | Thomas Seidler and Markus Abel | Transition of AI Models in dependence of noise | submitted to Physical Review E | null | null | null | physics.data-an | http://creativecommons.org/licenses/by-sa/4.0/ | We investigate the dependence of the score on noise in the data, and on the network size. As a result, we obtain the so-called "cognition transition" from good performance to zero with increasing noise. The understanding of this transition is of fundamental scientific and practical interest. We use concepts from statistical mechanics to understand how a changing finite size of models affects the cognition ability under the presence or corrupted data. On one hand, we study if there is a universal aspect in the transition to several models, on the other hand we go into detail how the approach of the cognition transition point can be captured quantitatively. Therefore, we use the so-called scaling approach from statistical mechanics and find a power-law behaviour of the transition width with increasing model size. Since our study is aimed at universal aspects we use well-know models and data for image classification. That way we avoid uncertainties in data handling or model setup. The practical implication of our results is a tool to estimate model sizes for a certain "universality class" of models, without the need to investigate large sizes, just by extrapolating the scaling results. In turn, that allows for cost reduction in hyperparameter studies. Here, we present first results on a concrete setup; we think that the understanding the mechanics of large system sizes is of fundamental interest for a further exploration of even larger models. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 03:19:29 GMT"
},
{
"version": "v2",
"created": "Mon, 30 Jun 2025 07:16:56 GMT"
},
{
"version": "v3",
"created": "Thu, 03 Jul 2025 08:43:51 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Seidler",
"Thomas",
""
],
[
"Abel",
"Markus",
""
]
] |
2506.16889 | Junghyun (Tony) Koo | Junghyun Koo, Marco A. Mart\'inez-Ram\'irez, Wei-Hsiang Liao, Giorgio Fabbro, Michele Mancusi, and Yuki Mitsufuji | ITO-Master: Inference-Time Optimization for Audio Effects Modeling of Music Mastering Processors | ISMIR 2025 | null | null | null | cs.SD eess.AS | http://creativecommons.org/licenses/by/4.0/ | Music mastering style transfer aims to model and apply the mastering characteristics of a reference track to a target track, simulating the professional mastering process. However, existing methods apply fixed processing based on a reference track, limiting users' ability to fine-tune the results to match their artistic intent. In this paper, we introduce the ITO-Master framework, a reference-based mastering style transfer system that integrates Inference-Time Optimization (ITO) to enable finer user control over the mastering process. By optimizing the reference embedding during inference, our approach allows users to refine the output dynamically, making micro-level adjustments to achieve more precise mastering results. We explore both black-box and white-box methods for modeling mastering processors and demonstrate that ITO improves mastering performance across different styles. Through objective evaluation, subjective listening tests, and qualitative analysis using text-based conditioning with CLAP embeddings, we validate that ITO enhances mastering style similarity while offering increased adaptability. Our framework provides an effective and user-controllable solution for mastering style transfer, allowing users to refine their results beyond the initial style transfer. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 10:21:56 GMT"
},
{
"version": "v2",
"created": "Thu, 26 Jun 2025 08:38:02 GMT"
},
{
"version": "v3",
"created": "Thu, 03 Jul 2025 03:36:09 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Koo",
"Junghyun",
""
],
[
"Martínez-Ramírez",
"Marco A.",
""
],
[
"Liao",
"Wei-Hsiang",
""
],
[
"Fabbro",
"Giorgio",
""
],
[
"Mancusi",
"Michele",
""
],
[
"Mitsufuji",
"Yuki",
""
]
] |
2506.17347 | Jennifer Wang | Jennifer Wang, Andrew Selbst, Solon Barocas, Suresh Venkatasubramanian | Distinguishing Predictive and Generative AI in Regulation | null | null | null | null | cs.CY cs.AI | http://creativecommons.org/licenses/by/4.0/ | Over the past decade, policymakers have developed a set of regulatory tools to ensure AI development aligns with key societal goals. Many of these tools were initially developed in response to concerns with predictive AI and therefore encode certain assumptions about the nature of AI systems and the utility of certain regulatory approaches. With the advent of generative AI, however, some of these assumptions no longer hold, even as policymakers attempt to maintain a single regulatory target that covers both types of AI. In this paper, we identify four distinct aspects of generative AI that call for meaningfully different policy responses. These are the generality and adaptability of generative AI that make it a poor regulatory target, the difficulty of designing effective evaluations, new legal concerns that change the ecosystem of stakeholders and sources of expertise, and the distributed structure of the generative AI value chain. In light of these distinctions, policymakers will need to evaluate where the past decade of policy work remains relevant and where new policies, designed to address the unique risks posed by generative AI, are necessary. We outline three recommendations for policymakers to more effectively identify regulatory targets and leverage constraints across the broader ecosystem to govern generative AI. | [
{
"version": "v1",
"created": "Thu, 19 Jun 2025 17:17:55 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 22:50:44 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Wang",
"Jennifer",
""
],
[
"Selbst",
"Andrew",
""
],
[
"Barocas",
"Solon",
""
],
[
"Venkatasubramanian",
"Suresh",
""
]
] |
2506.17365 | Motoyuki Nobori | Motoyuki Nobori | A Generalization of the B\"{o}ttcher-Wenzel inequality for three rectangular matrices | 9 pages | null | null | null | math.RA math.FA math.OA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Let $m,n$ be positive integers. For all $m\times n$ complex matrices $A, C$ and an $n\times m$ matrix $B$, we define a generalized commutator as $ABC-CBA$. We estimate the Frobenius norm of it, and finally get the inequality, which is a generalization of the B\"{o}ttcher-Wenzel inequality. If $n=1$ or $m=1$, then the Frobenius norm of $ABC-CBA$ can be estimated with a tighter upper bound. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 11:49:50 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 14:22:00 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Nobori",
"Motoyuki",
""
]
] |
2506.17423 | Yao Huang | Baolin Zhang, Zixiao Ma, Yao Huang, Huili Han, Ruming Hu, Yuzhuo Wang, Huaqing Zhang, Liyan Tang, Tingyun Shi, Hua Guan, Kelin Gao | A Liquid-Nitrogen-Cooled Ca+ Ion Optical Clock with a Systematic Uncertainty of 4.6E-19 | 12 pages, 14 figures | null | null | null | physics.atom-ph physics.optics | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We report a single-ion optical clock based on the 4S_1/2-3D_5/2 transition of the 40Ca+ ion, operated in a liquid nitrogen cryogenic environment,achieving a total systematic uncertainty of 4.6E-19. We employ a refined temperature evaluation scheme to reduce the frequency uncertainty due to blackbody radiation (BBR), and the 3D sideband cooling has been implemented to minimize the second-order Doppler shift. We have precisely determined the average Zeeman coefficient of the 40Ca+ clock transition to be 14.345(40) Hz/mT^2, thereby significantly reducing the quadratic Zeeman shift uncertainty. Moreover, the cryogenic environment enables the lowest reported heating rate due to ambient electric field noise in trapped-ion optical clocks. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 18:41:34 GMT"
},
{
"version": "v2",
"created": "Wed, 25 Jun 2025 10:06:30 GMT"
},
{
"version": "v3",
"created": "Thu, 03 Jul 2025 10:07:10 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Zhang",
"Baolin",
""
],
[
"Ma",
"Zixiao",
""
],
[
"Huang",
"Yao",
""
],
[
"Han",
"Huili",
""
],
[
"Hu",
"Ruming",
""
],
[
"Wang",
"Yuzhuo",
""
],
[
"Zhang",
"Huaqing",
""
],
[
"Tang",
"Liyan",
""
],
[
"Shi",
"Tingyun",
""
],
[
"Guan",
"Hua",
""
],
[
"Gao",
"Kelin",
""
]
] |
2506.17535 | Iv\'an M. \'Alvarez Rios | Iv\'an \'Alvarez-Rios, Carlos Tena-Contreras, and Francisco S. Guzm\'an | Kinematic Imprints of vortex-lines of BEC Dark Matter on Baryonic Matter | 12 pages, 12 figures | Phys. Rev. D 111, 123039 (2025) | 10.1103/x9mt-wprk | null | astro-ph.GA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Our results demonstrate that vortex lines in Bose-Einstein Condensate Dark Matter (BECDM) can act as gravitational seeds that induce the condensation of baryonic matter, leading to localized gas accumulation even in the absence of imposed symmetries or rotation. Our analysis is based on the numerical solution of the system of equations for the BECDM gravitationally coupled to Euler equations for a compressible ideal gas (IG) that we use as a model of baryonic matter. Numerical simulations are constructed for various scenarios that start with a vortex solution for the BECDM and a randomly distributed ideal gas, with the aim of investigating whether the matter distribution and dynamics of the vortex influences the dynamics and distribution of the gas. We find that the gas condensation process is most efficient when the IG mass dominates over the BECDM, and when the IG has low initial velocity dispersion. We also find that strong bosonic self-interaction does not guarantee the vortex stability, instead, it can trigger dynamical instabilities that disrupt both the vortex structure and the surrounding gas. An interesting finding is that the vortex drives a persistent morphological signature on the gas, often in the form of ring-like features visible in projected density maps. These patterns survive nonlinear evolution and may serve as indirect tracers of vortex structures in BECDM halos, potentially offering a novel and testable observational probe of the model. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 01:12:53 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Álvarez-Rios",
"Iván",
""
],
[
"Tena-Contreras",
"Carlos",
""
],
[
"Guzmán",
"Francisco S.",
""
]
] |
2506.17701 | Chung-Jun Tsai | Chung-Jun Tsai, Mao-Pei Tsui, Mu-Tao Wang | An ansatz for constructing explicit solutions of Hessian equations | 25 pages. A subsection is added to discuss all "variations of quadrics solutions" to constant-coefficient Hessian equations when n=3 | null | null | null | math.DG math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce a (variation of quadrics) ansatz for constructing explicit, real-valued solutions to broad classes of complex Hessian equations on domains in $\mathbb{C}^{n+1}$ and real Hessian equations on domains in $\mathbb{R}^{n+1}$. In the complex setting, our method simultaneously addresses the deformed Hermitian--Yang--Mills/Leung--Yau--Zaslow (dHYM/LYZ) equation, the Monge--Amp\`ere equation, and the $J$-equation. Under this ansatz each PDE reduces to a second-order system of ordinary differential equations admitting explicit first integrals. These ODE systems integrate in closed form via abelian integrals, producing wide families of explicit solutions together with a detailed description. In particular, on $\mathbb{C}^3$, we construct entire dHYM/LYZ solutions of arbitrary subcritical phase, and on $\mathbb{R}^3$ we produce entire special Lagrangian solutions of arbitrary subcritical phase. More generally, in any complex or real dimension, our ansatz yields entire solutions of certain subcritical phases for both the dHYM/LYZ and special Lagrangian equations. Some of these solutions develop singularities on compact regions. In the special Lagrangian case we show that, after a natural extension across the singular locus, these blow-up solutions coincide with previously known complete special Lagrangian submanifolds obtained via a different ansatz. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 12:28:20 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 06:27:36 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Tsai",
"Chung-Jun",
""
],
[
"Tsui",
"Mao-Pei",
""
],
[
"Wang",
"Mu-Tao",
""
]
] |
2506.17754 | Dongzhe Zheng | Dongzhe Zheng | Exceptional Symmetry as a Source of Algebraic Cycles: Non-Constructive Methods for the Hodge Conjecture for a Special Class of Calabi-Yau 5-Folds | null | null | null | null | math.AG math-ph math.DG math.MP math.RT | http://creativecommons.org/licenses/by/4.0/ | Classical variational Hodge structure theory characterizes the algebraicity of Hodge classes by studying the transversality of period mappings under geometric deformations. However, when algebraic varieties lack appropriate deformation families, this method faces applicability limitations. This paper develops a non-constructive method based on exceptional Lie group constraints to handle such cases. Our main technical contribution is establishing a dimension control mechanism for Spencer cohomology theory under Lie group constraints. Specifically, we prove that when a compact K\"ahler manifold $X$ is equipped with $E_7$ group constraints, the corresponding Spencer kernel $\mathcal{K}_\lambda^{1,1}$ has complex dimension simultaneously constrained by two bounds: representation theory gives the lower bound $\dim_{\mathbb{C}}\mathcal{K}_\lambda^{1,1} \geq 56$, while the degenerate Spencer-de Rham mapping gives the upper bound $\dim_{\mathbb{C}}\mathcal{K}_\lambda^{1,1} \leq h^{1,1}(X)$. For 5-dimensional Calabi-Yau manifolds satisfying $h^{1,1}(X) = 56$, this dimension constraint becomes the equality $\dim_{\mathbb{C}}\mathcal{K}_\lambda^{1,1} = 56 = h^{1,1}(X)$. Combined with our established Spencer-calibration equivalence principle, this dimension matching result is sufficient to verify the $(1,1)$-type Hodge conjecture. Our proof completely avoids explicit algebraic cycle construction, instead achieving the goal through abstract dimensional arguments. This method demonstrates the application potential of Lie group representation theory in algebraic geometry, providing new theoretical tools for handling geometric objects where traditional deformation methods are not applicable. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 16:18:48 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 18:38:00 GMT"
},
{
"version": "v3",
"created": "Thu, 03 Jul 2025 02:07:10 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Zheng",
"Dongzhe",
""
]
] |
2506.17827 | Michele Castellana | Dennis W\"orthm\"uller, Gaetano Ferraro, Pierre Sens, Michele Castellana | IRENE: a fluId layeR finitE-elemeNt softwarE | null | null | null | null | physics.flu-dyn physics.comp-ph | http://creativecommons.org/licenses/by/4.0/ | We present a finite-element software library, IRENE, for numerically solving the steady state and dynamics of a two-dimensional viscous fluid layer embedded in three-dimensional space, on multiple physical scales. The library is designed to handle a wide range of physical regimes -- both low-Reynolds-number and turbulent ones -- and geometries, capturing the coupling between in-plane flows, out-of-plane deformations, surface tension, and elastic response. In addition, the software can treat complex geometries, including those with intra-layer obstacles or arbitrary layer boundaries. We validate IRENE against known analytical and numerical results and demonstrate its capabilities through examples relevant to biological membranes and macroscopic air flows. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 21:44:10 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 15:42:47 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Wörthmüller",
"Dennis",
""
],
[
"Ferraro",
"Gaetano",
""
],
[
"Sens",
"Pierre",
""
],
[
"Castellana",
"Michele",
""
]
] |
2506.17828 | Xinnan Zhang | Xinnan Zhang, Chenliang Li, Siliang Zeng, Jiaxiang Li, Zhongruo Wang, Kaixiang Lin, Songtao Lu, Alfredo Garcia, Mingyi Hong | Aligning Frozen LLMs by Reinforcement Learning: An Iterative Reweight-then-Optimize Approach | null | null | null | null | cs.LG cs.AI cs.CL | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Aligning large language models (LLMs) with human preferences usually requires fine-tuning methods such as RLHF and DPO. These methods directly optimize the model parameters, so they cannot be used in test-time to improve model performance, nor are they applicable when the model weights are not accessible. In contrast, test-time methods sidestep weight updates by leveraging reward functions to guide and improve output quality. However, they incur high inference costs, and their one-shot guidance is often based on imperfect reward or value functions, leading to suboptimal outputs. In this work, we present a method named Iterative Reweight-then-Optimize (IRO), a reinforcement learning (RL) framework that performs RL-style alignment of the (frozen) base model without touching its parameters. During training, each iteration (i) samples candidates from the base model, (ii) resamples using current value functions, and (iii) trains a new lightweight value function that guides the next decoding pass. At test time, the value functions are used to guide the base model generation via a search-based optimization process. Notably, users can apply IRO to align a model on their own dataset, similar to OpenAI's reinforcement fine-tuning (RFT), but without requiring access to the model weights. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 21:49:02 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 05:12:51 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Zhang",
"Xinnan",
""
],
[
"Li",
"Chenliang",
""
],
[
"Zeng",
"Siliang",
""
],
[
"Li",
"Jiaxiang",
""
],
[
"Wang",
"Zhongruo",
""
],
[
"Lin",
"Kaixiang",
""
],
[
"Lu",
"Songtao",
""
],
[
"Garcia",
"Alfredo",
""
],
[
"Hong",
"Mingyi",
""
]
] |
2506.17915 | Cheng Zeng | Cheng Zeng, Gengji Li | Some sharp bounds on the average Steiner (k, l)-eccentricity for trees | 14 pages, 6 figures | null | null | null | math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper we introduce some transformations for trees that do not increase the average Steiner $(k,l)$-eccentricity for all $0\leq l\leq k\leq n$. Using these transformations, we obtain some sharp bounds on the average Steiner $(k,l)$-eccentricity for trees with some certain conditions, including given nodes, given diameter, given max degree and given leaves, and get the corresponding extremal trees as well. | [
{
"version": "v1",
"created": "Sun, 22 Jun 2025 06:34:59 GMT"
},
{
"version": "v2",
"created": "Wed, 25 Jun 2025 12:06:05 GMT"
},
{
"version": "v3",
"created": "Thu, 26 Jun 2025 13:28:12 GMT"
},
{
"version": "v4",
"created": "Sat, 28 Jun 2025 09:34:58 GMT"
},
{
"version": "v5",
"created": "Thu, 03 Jul 2025 03:08:50 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Zeng",
"Cheng",
""
],
[
"Li",
"Gengji",
""
]
] |
2506.18248 | Jongoh Jeong | Jongoh Jeong, Hunmin Yang, Jaeseok Jeong, Kuk-Jin Yoon | Semantic Structure-Aware Generative Attacks for Enhanced Adversarial Transferability | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Generative adversarial attacks train a perturbation generator on a white-box surrogate model and subsequently apply the crafted perturbations to unseen black-box victim models. In contrast to iterative attacks, these methods deliver superior inference-time efficiency, scalability, and transferability; however, up until now, existing studies have not fully exploited the representational capacity of generative models to preserve and harness semantic information. Specifically, the intermediate activations of the generator encode rich semantic features--object boundaries and coarse shapes--that remain under-exploited, thereby limiting the alignment of perturbations with object-salient regions which are critical for adversarial transferability. To remedy this, we introduce a semantic structure-aware attack framework based on the Mean Teacher, which serves as a temporally smoothed feature reference. With this smoothed reference, we further direct semantic consistency between the early-layer activations in the student and those of the semantically rich teacher by feature distillation. By anchoring perturbation synthesis to the semantically salient early intermediate blocks within the generator based on empirical findings, our method guides progressive adversarial perturbation on regions that substantially enhance adversarial transferability. We conduct extensive experiments over diverse models, domains and tasks to demonstrate consistent improvements relative to state-of-the-art generative attacks, comprehensively evaluated using conventional metrics and our newly proposed Accidental Correction Rate (ACR). | [
{
"version": "v1",
"created": "Mon, 23 Jun 2025 02:35:09 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 03:17:52 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Jeong",
"Jongoh",
""
],
[
"Yang",
"Hunmin",
""
],
[
"Jeong",
"Jaeseok",
""
],
[
"Yoon",
"Kuk-Jin",
""
]
] |
2506.18435 | Alessandro Daniele Genuardi Oquendo | Alessandro Daniele Genuardi Oquendo, Ali Nadir, Tigers Jonuzi, Siddhartha Patra, Nilotpal Kanti Sinha, Rom\'an Or\'us, Sam Mugel | Accelerating Photonic Integrated Circuit Design: Traditional, ML and Quantum Methods | We thank the Centro para el Desarrollo Tecnol\'ogico e Industria (CDTI) for their support under the Project INSPIRE via Misiones PERTE Chip 2023. Grant EXP-00162621 / MIG-20231016 | null | null | null | quant-ph physics.app-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Photonic Integrated Circuits (PICs) provide superior speed, bandwidth, and energy efficiency, making them ideal for communication, sensing, and quantum computing applications. Despite their potential, PIC design workflows and integration lag behind those in electronics, calling for groundbreaking advancements. This review outlines the state of PIC design, comparing traditional simulation methods with machine learning approaches that enhance scalability and efficiency. It also explores the promise of quantum algorithms and quantum-inspired methods to address design challenges. | [
{
"version": "v1",
"created": "Mon, 23 Jun 2025 09:17:22 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 10:56:38 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Oquendo",
"Alessandro Daniele Genuardi",
""
],
[
"Nadir",
"Ali",
""
],
[
"Jonuzi",
"Tigers",
""
],
[
"Patra",
"Siddhartha",
""
],
[
"Sinha",
"Nilotpal Kanti",
""
],
[
"Orús",
"Román",
""
],
[
"Mugel",
"Sam",
""
]
] |
2506.18482 | Leonard S. Pleiss | Leonard S. Pleiss, Tobias Sutter, Maximilian Schiffer | Reliability-Adjusted Prioritized Experience Replay | null | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Experience replay enables data-efficient learning from past experiences in online reinforcement learning agents. Traditionally, experiences were sampled uniformly from a replay buffer, regardless of differences in experience-specific learning potential. In an effort to sample more efficiently, researchers introduced Prioritized Experience Replay (PER). In this paper, we propose an extension to PER by introducing a novel measure of temporal difference error reliability. We theoretically show that the resulting transition selection algorithm, Reliability-adjusted Prioritized Experience Replay (ReaPER), enables more efficient learning than PER. We further present empirical results showing that ReaPER outperforms PER across various environment types, including the Atari-10 benchmark. | [
{
"version": "v1",
"created": "Mon, 23 Jun 2025 10:35:36 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 09:55:57 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Pleiss",
"Leonard S.",
""
],
[
"Sutter",
"Tobias",
""
],
[
"Schiffer",
"Maximilian",
""
]
] |
2506.18549 | Paolo Ferro | Paolo Ferro | Bounded information as a foundation for quantum theory | 73 pages, 11 figures | null | null | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | The purpose of this paper is to formalize the concept that best synthesizes our intuitive understanding of quantum mechanics -- that the information carried by a system is limited -- and, from this principle, to construct the foundations of quantum theory. In our discussion, we also introduce a second important hypothesis: if a measurement closely approximates an ideal one in terms of experimental precision, the information it provides about a physical system is independent of the measurement method and, specifically, of the system's physical quantities being measured. This principle can be expressed in terms of metric properties of a manifold whose points represent the state of the system. These and other reasonable hypotheses provide the foundation for a framework of quantum reconstruction.
The theory presented in this paper is based on a description of physical systems in terms of their statistical properties, specifically statistical parameters, and focuses on the study of estimators for these parameters. To achieve the goal of quantum reconstruction, a divide-and-conquer approach is employed, wherein the space of two discrete conjugate Hamiltonian variables is partitioned into a binary tree of nested sets. This approach naturally leads to the reconstruction of the linear and probabilistic structure of quantum mechanics. | [
{
"version": "v1",
"created": "Mon, 23 Jun 2025 11:57:26 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 16:19:27 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Ferro",
"Paolo",
""
]
] |
2506.18803 | Pat Lank | Pat Lank | Perfectly generated $t$-structures for algebraic stacks | Current: Refinement to qcqs case and details added. Previous: Upgrade to qcqs stacks by using only Hrbek's affine result | null | null | null | math.AG math.AC | http://creativecommons.org/licenses/by/4.0/ | This work studies $t$-structures for the derived category of quasi-coherent sheaves on a quasi-compact quasi-separated algebraic stack. Specifically, using Thomason filtrations, we classify those $t$-structures which are generated by perfect complexes and satisfy a tensor compatibility. Interestingly, the only input required is the classification \`{a} la Hrbek for the affine scheme case. | [
{
"version": "v1",
"created": "Mon, 23 Jun 2025 16:07:49 GMT"
},
{
"version": "v2",
"created": "Wed, 25 Jun 2025 17:04:00 GMT"
},
{
"version": "v3",
"created": "Thu, 03 Jul 2025 15:39:47 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Lank",
"Pat",
""
]
] |
2506.18820 | Hirakjyoti Sarma | Hirakjyoti Sarma and Nilakshi Das | Structure and dynamics of finite three-dimensional Yukawa clusters in complex plasmas : Newtonian versus Langevin Dynamics | 26 pages, 24 figures | null | null | null | physics.plasm-ph cond-mat.stat-mech | http://creativecommons.org/licenses/by/4.0/ | The structure and dynamics of a harmonically confined three dimensional finite dust cluster are investigated via both Langevin Dynamics (LD) and frictionless Molecular Dynamics (fMD) simulation. The static structure of the system is analyzed through the Radial Distribution Function, Center-two-particle correlation function(C2P) and angular correlation function. The intra-shell angular correlation, Radial Distribution Function and C2P remains largely unaffected by the dynamics employed. However, the inter-shell angular correlation exhibits sharp peaks at irregular angular intervals in fMD which are not seen in LD indicating strongly correlated motion of the two spherical shells in the cluster in fMD. The single particle dynamics of the cluster is characterized by the Van - Hove self autocorrelation function and Mean Squared Displacement (MSD). Notably, the Van - Hove autocorrelation function in fMD simulations exhibits narrower width and higher peak height as compared to the LD simulations, suggesting greater particle mobility in LD. Trajectory analysis reveals a rotational motion of the particles about a common axis in fMD which disappears with progressively increasing friction coefficient. We show that the disappearance of rotational motion with the introduction of neutral friction in the dynamics is due to the much faster relaxation of the interparticle distance as well as interparticle angular separation in LD as compared to fMD. | [
{
"version": "v1",
"created": "Mon, 23 Jun 2025 16:28:04 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 04:18:17 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Sarma",
"Hirakjyoti",
""
],
[
"Das",
"Nilakshi",
""
]
] |
2506.18959 | Weizhi Zhang | Weizhi Zhang, Yangning Li, Yuanchen Bei, Junyu Luo, Guancheng Wan, Liangwei Yang, Chenxuan Xie, Yuyao Yang, Wei-Chieh Huang, Chunyu Miao, Henry Peng Zou, Xiao Luo, Yusheng Zhao, Yankai Chen, Chunkit Chan, Peilin Zhou, Xinyang Zhang, Chenwei Zhang, Jingbo Shang, Ming Zhang, Yangqiu Song, Irwin King, Philip S. Yu | From Web Search towards Agentic Deep Research: Incentivizing Search with Reasoning Agents | null | null | null | null | cs.IR cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Information retrieval is a cornerstone of modern knowledge acquisition, enabling billions of queries each day across diverse domains. However, traditional keyword-based search engines are increasingly inadequate for handling complex, multi-step information needs. Our position is that Large Language Models (LLMs), endowed with reasoning and agentic capabilities, are ushering in a new paradigm termed Agentic Deep Research. These systems transcend conventional information search techniques by tightly integrating autonomous reasoning, iterative retrieval, and information synthesis into a dynamic feedback loop. We trace the evolution from static web search to interactive, agent-based systems that plan, explore, and learn. We also introduce a test-time scaling law to formalize the impact of computational depth on reasoning and search. Supported by benchmark results and the rise of open-source implementations, we demonstrate that Agentic Deep Research not only significantly outperforms existing approaches, but is also poised to become the dominant paradigm for future information seeking. All the related resources, including industry products, research papers, benchmark datasets, and open-source implementations, are collected for the community in https://github.com/DavidZWZ/Awesome-Deep-Research. | [
{
"version": "v1",
"created": "Mon, 23 Jun 2025 17:27:19 GMT"
},
{
"version": "v2",
"created": "Thu, 26 Jun 2025 17:18:00 GMT"
},
{
"version": "v3",
"created": "Thu, 03 Jul 2025 17:48:36 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Zhang",
"Weizhi",
""
],
[
"Li",
"Yangning",
""
],
[
"Bei",
"Yuanchen",
""
],
[
"Luo",
"Junyu",
""
],
[
"Wan",
"Guancheng",
""
],
[
"Yang",
"Liangwei",
""
],
[
"Xie",
"Chenxuan",
""
],
[
"Yang",
"Yuyao",
""
],
[
"Huang",
"Wei-Chieh",
""
],
[
"Miao",
"Chunyu",
""
],
[
"Zou",
"Henry Peng",
""
],
[
"Luo",
"Xiao",
""
],
[
"Zhao",
"Yusheng",
""
],
[
"Chen",
"Yankai",
""
],
[
"Chan",
"Chunkit",
""
],
[
"Zhou",
"Peilin",
""
],
[
"Zhang",
"Xinyang",
""
],
[
"Zhang",
"Chenwei",
""
],
[
"Shang",
"Jingbo",
""
],
[
"Zhang",
"Ming",
""
],
[
"Song",
"Yangqiu",
""
],
[
"King",
"Irwin",
""
],
[
"Yu",
"Philip S.",
""
]
] |
2506.19181 | Xin Zhu | Xin Zhu, Ahmet Enis Cetin, Gorkem Durak, Batuhan Gundogdu, Ziliang Hong, Hongyi Pan, Ertugrul Aktas, Elif Keles, Hatice Savas, Aytekin Oto, Hiten Patel, Adam B. Murphy, Ashley Ross, Frank Miller, Baris Turkbey, Ulas Bagci | VHU-Net: Variational Hadamard U-Net for Body MRI Bias Field Correction | null | null | null | null | eess.IV | http://creativecommons.org/licenses/by/4.0/ | Bias field artifacts in magnetic resonance imaging (MRI) scans introduce spatially smooth intensity inhomogeneities that degrade image quality and hinder downstream analysis. To address this challenge, we propose a novel variational Hadamard U-Net (VHU-Net) for effective body MRI bias field correction. The encoder comprises multiple convolutional Hadamard transform blocks (ConvHTBlocks), each integrating convolutional layers with a Hadamard transform (HT) layer. Specifically, the HT layer performs channel-wise frequency decomposition to isolate low-frequency components, while a subsequent scaling layer and semi-soft thresholding mechanism suppress redundant high-frequency noise. To compensate for the HT layer's inability to model inter-channel dependencies, the decoder incorporates an inverse HT-reconstructed transformer block, enabling global, frequency-aware attention for the recovery of spatially consistent bias fields. The stacked decoder ConvHTBlocks further enhance the capacity to reconstruct the underlying ground-truth bias field. Building on the principles of variational inference, we formulate a new evidence lower bound (ELBO) as the training objective, promoting sparsity in the latent space while ensuring accurate bias field estimation. Comprehensive experiments on abdominal and prostate MRI datasets demonstrate the superiority of VHU-Net over existing state-of-the-art methods in terms of intensity uniformity, signal fidelity, and tissue contrast. Moreover, the corrected images yield substantial downstream improvements in segmentation accuracy. Our framework offers computational efficiency, interpretability, and robust performance across multi-center datasets, making it suitable for clinical deployment. | [
{
"version": "v1",
"created": "Mon, 23 Jun 2025 22:57:38 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 20:30:31 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Zhu",
"Xin",
""
],
[
"Cetin",
"Ahmet Enis",
""
],
[
"Durak",
"Gorkem",
""
],
[
"Gundogdu",
"Batuhan",
""
],
[
"Hong",
"Ziliang",
""
],
[
"Pan",
"Hongyi",
""
],
[
"Aktas",
"Ertugrul",
""
],
[
"Keles",
"Elif",
""
],
[
"Savas",
"Hatice",
""
],
[
"Oto",
"Aytekin",
""
],
[
"Patel",
"Hiten",
""
],
[
"Murphy",
"Adam B.",
""
],
[
"Ross",
"Ashley",
""
],
[
"Miller",
"Frank",
""
],
[
"Turkbey",
"Baris",
""
],
[
"Bagci",
"Ulas",
""
]
] |
2506.19283 | Xiangbo Gao | Xiangbo Gao, Yuheng Wu, Fengze Yang, Xuewen Luo, Keshu Wu, Xinghao Chen, Yuping Wang, Chenxi Liu, Yang Zhou, Zhengzhong Tu | AirV2X: Unified Air-Ground Vehicle-to-Everything Collaboration | null | null | null | null | cs.CV cs.AI cs.RO | http://creativecommons.org/licenses/by/4.0/ | While multi-vehicular collaborative driving demonstrates clear advantages over single-vehicle autonomy, traditional infrastructure-based V2X systems remain constrained by substantial deployment costs and the creation of "uncovered danger zones" in rural and suburban areas. We present AirV2X-Perception, a large-scale dataset that leverages Unmanned Aerial Vehicles (UAVs) as a flexible alternative or complement to fixed Road-Side Units (RSUs). Drones offer unique advantages over ground-based perception: complementary bird's-eye-views that reduce occlusions, dynamic positioning capabilities that enable hovering, patrolling, and escorting navigation rules, and significantly lower deployment costs compared to fixed infrastructure. Our dataset comprises 6.73 hours of drone-assisted driving scenarios across urban, suburban, and rural environments with varied weather and lighting conditions. The AirV2X-Perception dataset facilitates the development and standardized evaluation of Vehicle-to-Drone (V2D) algorithms, addressing a critical gap in the rapidly expanding field of aerial-assisted autonomous driving systems. The dataset and development kits are open-sourced at https://github.com/taco-group/AirV2X-Perception. | [
{
"version": "v1",
"created": "Tue, 24 Jun 2025 03:34:39 GMT"
},
{
"version": "v2",
"created": "Tue, 01 Jul 2025 01:20:24 GMT"
},
{
"version": "v3",
"created": "Wed, 02 Jul 2025 23:07:24 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Gao",
"Xiangbo",
""
],
[
"Wu",
"Yuheng",
""
],
[
"Yang",
"Fengze",
""
],
[
"Luo",
"Xuewen",
""
],
[
"Wu",
"Keshu",
""
],
[
"Chen",
"Xinghao",
""
],
[
"Wang",
"Yuping",
""
],
[
"Liu",
"Chenxi",
""
],
[
"Zhou",
"Yang",
""
],
[
"Tu",
"Zhengzhong",
""
]
] |
2506.19501 | Ziyao Xiong | Ziyao Xiong, Qixing Deng, Yidan Sun, Junhua Yang | B0 -> K*0 tau+ tau- Decay: Using Machine Learning to Separate Signal from Background | This submission is being withdrawn because it contains experimental data that I included without obtaining the necessary authorization. The data appears in the Results and Discussion section and was not approved for public release | null | null | null | hep-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This study investigates the rare decay B0 -> K*0 tau+ tau-, which is sensitive to potential violations of lepton flavor universality predicted by the Standard Model. A Monte Carlo simulated dataset containing both signal and the dominant background process B0 -> K*0 D+ D- was used to train and evaluate machine learning classifiers. After feature selection and parameter tuning, two supervised models -- Boosted Decision Trees (BDTs) and Fully Connected Neural Networks (FCNNs) -- were trained. Feature engineering was then applied to enhance classification performance. On the test set, the BDT achieved an AUC of 0.912 +/- 0.000 and an F1-score of 0.828 +/- 0.001, while the FCNN reached an AUC of 0.877 +/- 0.000 and an F1-score of 0.799 +/- 0.001. These results demonstrate that both models can robustly separate signal from background in rare decay searches, supporting their application in future LHCb analyses. | [
{
"version": "v1",
"created": "Tue, 24 Jun 2025 10:40:16 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 02:05:34 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Xiong",
"Ziyao",
""
],
[
"Deng",
"Qixing",
""
],
[
"Sun",
"Yidan",
""
],
[
"Yang",
"Junhua",
""
]
] |
2506.19660 | Shuhang Xu | Shuhang Xu, Yunfei Gu, Linhui Liu, Chentao Wu | PS-WL: A Probability-Sensitive Wear Leveling scheme for SSD array scaling | null | null | null | null | cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As flash-based Solid State Drive (SSD) arrays become essential to modern data centers, scaling these arrays to meet explosive data growth is a frequent and critical operation. However, the conventional wear-leveling (WL) paradigm applied during scaling suffers from a fundamental flaw: it ignores the non-linear relationship between wear and failure probability, potentially pushing the most vulnerable, aged disks towards premature failure. To address this critical issue at its root, we propose the Probability-Sensitive Wear Leveling (PS-WL) scheme, which shifts the optimization goal from balancing wear to directly balancing failure risk. At its core, PS-WL introduces an "effective lifetime" model derived from a realistic failure probability to more accurately assess disk lifetime. This model guides a PID controller for wear leveling operation, with a conservative zone minimizes performance overhead by restricting warm data migration. Comprehensive simulations validate the superiority of PS-WL over state-of-the-art methods. The results demonstrate that our approach significantly reduces performance overhead while, most critically, consistently and effectively lowering the aggregated array failure risk across diverse system configurations and workloads. This proves that by directly optimizing for reliability, PS-WL builds a scalable storage system that is, by design, fundamentally safer, more efficient, and more stable. | [
{
"version": "v1",
"created": "Tue, 24 Jun 2025 14:22:56 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 15:00:42 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Xu",
"Shuhang",
""
],
[
"Gu",
"Yunfei",
""
],
[
"Liu",
"Linhui",
""
],
[
"Wu",
"Chentao",
""
]
] |
2506.19737 | Shuige Liu | Shuige Liu, Gabriel Ziegler | Reasoning about Bounded Reasoning | null | null | null | null | econ.TH | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Interactive decision-making relies on strategic reasoning. Two prominent frameworks are (1) models of bounded reasoning, exemplified by level-$k$ models, which keep reasoning implicit, and (2) epistemic game theory, which makes reasoning explicit. We connect these approaches by "lifting" static complete-information games into incomplete-information settings where payoff types reflect players' reasoning depths as in level-$k$ models. We introduce downward rationalizability, defined via minimal belief restrictions capturing the basic idea common to level-$k$ models, to provide robust yet well-founded predictions in games where bounded reasoning matters. We then refine these belief restrictions to analyze the foundations of two seminal models of bounded reasoning: the classic level-$k$ model and the cognitive hierarchy model. Our findings shed light on the distinction between hard cognitive bounds on reasoning and beliefs about co-players' types. Furthermore, they offer insights into robustness issues relevant for market design. Thus, our approach unifies key level-$k$ models building on clear foundations of strategic reasoning stemming from epistemic game theory. | [
{
"version": "v1",
"created": "Tue, 24 Jun 2025 15:56:24 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 14:29:44 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Liu",
"Shuige",
""
],
[
"Ziegler",
"Gabriel",
""
]
] |
2506.20122 | Kenneth Blakey | Kenneth Blakey, Ciprian Mircea Bonciocat | Parameterized Lagrangian Floer homotopy | 31 pages; v2: typos fixed and reference added | null | null | null | math.SG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We construct the Lagrangian Floer homotopy type, in the exact setting, as a spectrum parameterized over the moduli space of Maslov data. Our primary motivation for this construction is to provide stronger lower bounds for (possibly degenerate) Lagrangian intersections in plumbings of cotangent bundles. | [
{
"version": "v1",
"created": "Wed, 25 Jun 2025 04:26:58 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 03:45:14 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Blakey",
"Kenneth",
""
],
[
"Bonciocat",
"Ciprian Mircea",
""
]
] |
2506.20589 | Jorge Torres G\'omez | Jorge Torres G\'omez and Pit Hofmann and Lisa Y. Debus and Osman Tugay Ba\c{s}aran and Sebastian Lotter and Roya Khanzadeh and Stefan Angerbauer and Bige Deniz Unluturk and Sergi Abadal and Werner Haselmayr and Frank H.P. Fitzek and Robert Schober and Falko Dressler | Communicating Smartly in Molecular Communication Environments: Neural Networks in the Internet of Bio-Nano Things | Paper submitted to IEEE Communications Surveys & Tutorials | null | null | null | eess.SP cs.ET q-bio.OT | http://creativecommons.org/licenses/by/4.0/ | Recent developments in the Internet of Bio-Nano Things (IoBNT) are laying the groundwork for innovative applications across the healthcare sector. Nanodevices designed to operate within the body, managed remotely via the internet, are envisioned to promptly detect and actuate on potential diseases. In this vision, an inherent challenge arises due to the limited capabilities of individual nanosensors; specifically, nanosensors must communicate with one another to collaborate as a cluster. Aiming to research the boundaries of the clustering capabilities, this survey emphasizes data-driven communication strategies in molecular communication (MC) channels as a means of linking nanosensors. Relying on the flexibility and robustness of machine learning (ML) methods to tackle the dynamic nature of MC channels, the MC research community frequently refers to neural network (NN) architectures. This interdisciplinary research field encompasses various aspects, including the use of NNs to facilitate communication in MC environments, their implementation at the nanoscale, explainable approaches for NNs, and dataset generation for training. Within this survey, we provide a comprehensive analysis of fundamental perspectives on recent trends in NN architectures for MC, the feasibility of their implementation at the nanoscale, applied explainable artificial intelligence (XAI) techniques, and the accessibility of datasets along with best practices for their generation. Additionally, we offer open-source code repositories that illustrate NN-based methods to support reproducible research for key MC scenarios. Finally, we identify emerging research challenges, such as robust NN architectures, biologically integrated NN modules, and scalable training strategies. | [
{
"version": "v1",
"created": "Wed, 25 Jun 2025 16:28:30 GMT"
},
{
"version": "v2",
"created": "Thu, 26 Jun 2025 15:09:32 GMT"
},
{
"version": "v3",
"created": "Thu, 03 Jul 2025 14:47:17 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Gómez",
"Jorge Torres",
""
],
[
"Hofmann",
"Pit",
""
],
[
"Debus",
"Lisa Y.",
""
],
[
"Başaran",
"Osman Tugay",
""
],
[
"Lotter",
"Sebastian",
""
],
[
"Khanzadeh",
"Roya",
""
],
[
"Angerbauer",
"Stefan",
""
],
[
"Unluturk",
"Bige Deniz",
""
],
[
"Abadal",
"Sergi",
""
],
[
"Haselmayr",
"Werner",
""
],
[
"Fitzek",
"Frank H. P.",
""
],
[
"Schober",
"Robert",
""
],
[
"Dressler",
"Falko",
""
]
] |
2506.20907 | Juliana Garc\'ia-Mej\'ia | Juliana Garc\'ia-Mej\'ia, Zo\"e L. de Beurs, Patrick Tamburo, Andrew Vanderburg, David Charbonneau, Karen A. Collins, Khalid Barkaoui, Cristilyn N. Watkins, Chris Stockdale, Richard P. Schwarz, Raquel For\'es-Toribio, Jose A. Mu\~noz, Giovanni Isopi, Franco Mallia, Aldo Zapparata, Adam Popowicz, Andrzej Brudny, Eric Agol, Munazza K. Alam, Zouhair Benkhaldoun, Jehin Emmanuel, Mourad Ghachoui, Micha\"el Gillon, Keith Horne, Enric Pall\'e, Ramotholo Sefako, Avi Shporer, Mathilde Timmermans | A Ground-Based Transit Observation of the Long-Period Extremely Low-Density Planet HIP 41378 f | 14 pages, 5 figures. Submitted Wednesday June 18, 2025 to AAS Journals. v2: Updated to fix typo | null | null | null | astro-ph.EP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a ground-based transit detection of HIP 41378 f, a long-period ($P = 542$ days), extremely low-density ($0.09 \pm 0.02$ g cm$^{-3}$) giant exoplanet in a dynamically complex system. Using photometry from Tierras, TRAPPIST-North, and multiple LCOGT sites, we constrain the transit center time to $T_{C,6} = 2460438.889 \pm 0.049$ BJD TDB. This marks only the second ground-based detection of HIP 41378 f, currently the longest-period and longest-duration transiting exoplanet observed from the ground. We use this new detection to update the TTV solution for HIP 41378 f and refine the predicted times of its next two transits in November 2025 and April 2027. Incorporating new TESS Sector 88 data, we also rule out the 101-day orbital period alias for HIP 41378 d, and find that the remaining viable solutions are centered on the 278, 371, and 1113-day aliases. The latter two imply dynamical configurations that challenge the canonical view of planet e as the dominant perturber of planet f. Our results suggest that HIP 41378 d may instead play the leading role in shaping the TTV of HIP 41378 f. | [
{
"version": "v1",
"created": "Thu, 26 Jun 2025 00:25:39 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 19:28:33 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"García-Mejía",
"Juliana",
""
],
[
"de Beurs",
"Zoë L.",
""
],
[
"Tamburo",
"Patrick",
""
],
[
"Vanderburg",
"Andrew",
""
],
[
"Charbonneau",
"David",
""
],
[
"Collins",
"Karen A.",
""
],
[
"Barkaoui",
"Khalid",
""
],
[
"Watkins",
"Cristilyn N.",
""
],
[
"Stockdale",
"Chris",
""
],
[
"Schwarz",
"Richard P.",
""
],
[
"Forés-Toribio",
"Raquel",
""
],
[
"Muñoz",
"Jose A.",
""
],
[
"Isopi",
"Giovanni",
""
],
[
"Mallia",
"Franco",
""
],
[
"Zapparata",
"Aldo",
""
],
[
"Popowicz",
"Adam",
""
],
[
"Brudny",
"Andrzej",
""
],
[
"Agol",
"Eric",
""
],
[
"Alam",
"Munazza K.",
""
],
[
"Benkhaldoun",
"Zouhair",
""
],
[
"Emmanuel",
"Jehin",
""
],
[
"Ghachoui",
"Mourad",
""
],
[
"Gillon",
"Michaël",
""
],
[
"Horne",
"Keith",
""
],
[
"Pallé",
"Enric",
""
],
[
"Sefako",
"Ramotholo",
""
],
[
"Shporer",
"Avi",
""
],
[
"Timmermans",
"Mathilde",
""
]
] |
2506.21082 | Fei Ren | Fei Ren | Coherent six-functor formalisms: Pro vs Solid | 14 pages. v2: Prop 2.2 updated. Comments welcome! | null | null | null | math.AG math.GN | http://creativecommons.org/licenses/by/4.0/ | In the classical theory for coherent sheaves, the only missing piece in the Grothendieck six-functor formalism picture is $j_!$ for an open immersion $j$. Towards fixing this gap,classically Deligne proposed a construction of $j_!$ by extending the sheaf class to pro sheaves, while Clausen-Scholze provided another solution by extending the sheaf class to solid modules. In this work, we prove that Deligne's construction coincides with the Clausen-Scholze construction via a natural functor, whose restriction to the full subcategory of Mittag-Leffler pro-systems is fully faithful. | [
{
"version": "v1",
"created": "Thu, 26 Jun 2025 08:16:51 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 11:55:23 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Ren",
"Fei",
""
]
] |
2506.21191 | Koji Inoue | Koji Inoue, Mikey Elmers, Yahui Fu, Zi Haur Pang, Divesh Lala, Keiko Ochi, Tatsuya Kawahara | Prompt-Guided Turn-Taking Prediction | This paper has been accepted for presentation at SIGdial Meeting on Discourse and Dialogue 2025 (SIGDIAL 2025) and represents the author's version of the work | null | null | null | cs.CL cs.SD eess.AS | http://creativecommons.org/licenses/by/4.0/ | Turn-taking prediction models are essential components in spoken dialogue systems and conversational robots. Recent approaches leverage transformer-based architectures to predict speech activity continuously and in real-time. In this study, we propose a novel model that enables turn-taking prediction to be dynamically controlled via textual prompts. This approach allows intuitive and explicit control through instructions such as "faster" or "calmer" adapting dynamically to conversational partners and contexts. The proposed model builds upon a transformer-based voice activity projection (VAP) model, incorporating textual prompt embeddings into both channel-wise transformers and a cross-channel transformer. We evaluated the feasibility of our approach using over 950 hours of human-human spoken dialogue data. Since textual prompt data for the proposed approach was not available in existing datasets, we utilized a large language model (LLM) to generate synthetic prompt sentences. Experimental results demonstrated that the proposed model improved prediction accuracy and effectively varied turn-taking timing behaviors according to the textual prompts. | [
{
"version": "v1",
"created": "Thu, 26 Jun 2025 12:49:07 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 04:01:44 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Inoue",
"Koji",
""
],
[
"Elmers",
"Mikey",
""
],
[
"Fu",
"Yahui",
""
],
[
"Pang",
"Zi Haur",
""
],
[
"Lala",
"Divesh",
""
],
[
"Ochi",
"Keiko",
""
],
[
"Kawahara",
"Tatsuya",
""
]
] |
2506.21323 | Torben Skrzypek | Dennis le Plat, Torben Skrzypek | Three-point functions from integrability in $\mathcal{N}=2$ orbifold theories | 36 pages, 5 figures, 122 structure constants; v2: References added | null | null | DESY-25-088 | hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Besides solving the spectral problem of $\mathcal{N}=4$ Super-Yang-Mills (SYM) theory, integrability also provides us with tools to compute the structure constants of the theory, most prominently through the hexagon formalism. We show that, with minor modifications, this formalism can also be applied to orbifolds of $\mathcal{N}=4$ SYM theory, which are integrable theories in their own right. To substantiate this claim, we test our results against a direct gauge-theory calculation at tree-level. We focus here on a family of $\mathcal{N}=2$ supersymmetric $\mathbb{Z}_M$-orbifold theories. BPS correlators in these theories have recently been investigated with independent localisation techniques and a tentative matching with wrapping corrections in the hexagon formalism was observed. Together with our weak-coupling evidence, this suggests that a full determination of the structure constants of orbifold theories at finite coupling may be within reach. | [
{
"version": "v1",
"created": "Thu, 26 Jun 2025 14:39:10 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 09:15:00 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Plat",
"Dennis le",
""
],
[
"Skrzypek",
"Torben",
""
]
] |
2506.21506 | Boyu Gou | Boyu Gou, Zanming Huang, Yuting Ning, Yu Gu, Michael Lin, Weijian Qi, Andrei Kopanev, Botao Yu, Bernal Jim\'enez Guti\'errez, Yiheng Shu, Chan Hee Song, Jiaman Wu, Shijie Chen, Hanane Nour Moussa, Tianshu Zhang, Jian Xie, Yifei Li, Tianci Xue, Zeyi Liao, Kai Zhang, Boyuan Zheng, Zhaowei Cai, Viktor Rozgic, Morteza Ziyadi, Huan Sun, Yu Su | Mind2Web 2: Evaluating Agentic Search with Agent-as-a-Judge | Project Homepage: https://osu-nlp-group.github.io/Mind2Web-2/ | null | null | null | cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Agentic search such as Deep Research systems-where agents autonomously browse the web, synthesize information, and return comprehensive citation-backed answers-represents a major shift in how users interact with web-scale information. While promising greater efficiency and cognitive offloading, the growing complexity and open-endedness of agentic search have outpaced existing evaluation benchmarks and methodologies, which largely assume short search horizons and static answers. In this paper, we introduce Mind2Web 2, a benchmark of 130 realistic, high-quality, and long-horizon tasks that require real-time web browsing and extensive information synthesis, constructed with over 1000 hours of human labor. To address the challenge of evaluating time-varying and complex answers, we propose a novel Agent-as-a-Judge framework. Our method constructs task-specific judge agents based on a tree-structured rubric design to automatically assess both answer correctness and source attribution. We conduct a comprehensive evaluation of ten frontier agentic search systems and human performance, along with a detailed error analysis to draw insights for future development. The best-performing system, OpenAI Deep Research, can already achieve 50-70% of human performance while spending half the time, highlighting its great potential. Altogether, Mind2Web 2 provides a rigorous foundation for developing and benchmarking the next generation of agentic search systems. | [
{
"version": "v1",
"created": "Thu, 26 Jun 2025 17:32:50 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 15:47:40 GMT"
}
] | 2025-07-04T00:00:00 | [
[
"Gou",
"Boyu",
""
],
[
"Huang",
"Zanming",
""
],
[
"Ning",
"Yuting",
""
],
[
"Gu",
"Yu",
""
],
[
"Lin",
"Michael",
""
],
[
"Qi",
"Weijian",
""
],
[
"Kopanev",
"Andrei",
""
],
[
"Yu",
"Botao",
""
],
[
"Gutiérrez",
"Bernal Jiménez",
""
],
[
"Shu",
"Yiheng",
""
],
[
"Song",
"Chan Hee",
""
],
[
"Wu",
"Jiaman",
""
],
[
"Chen",
"Shijie",
""
],
[
"Moussa",
"Hanane Nour",
""
],
[
"Zhang",
"Tianshu",
""
],
[
"Xie",
"Jian",
""
],
[
"Li",
"Yifei",
""
],
[
"Xue",
"Tianci",
""
],
[
"Liao",
"Zeyi",
""
],
[
"Zhang",
"Kai",
""
],
[
"Zheng",
"Boyuan",
""
],
[
"Cai",
"Zhaowei",
""
],
[
"Rozgic",
"Viktor",
""
],
[
"Ziyadi",
"Morteza",
""
],
[
"Sun",
"Huan",
""
],
[
"Su",
"Yu",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.