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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2509.03315
|
Ruth Keogh
|
Ruth H. Keogh, Karla Diaz-Ordaz, Nan van Geloven, Jon Michael Gran, Kamaryn T. Tanner
|
The super learner for time-to-event outcomes: A tutorial
| null | null | null | null |
stat.ME
|
http://creativecommons.org/licenses/by/4.0/
|
Estimating risks or survival probabilities conditional on individual characteristics based on censored time-to-event data is a commonly faced task. This may be for the purpose of developing a prediction model or may be part of a wider estimation procedure, such as in causal inference. A challenge is that it is impossible to know at the outset which of a set of candidate models will provide the best predictions. The super learner is a powerful approach for finding the best model or combination of models ('ensemble') among a pre-specified set of candidate models or 'learners', which can include parametric and machine learning models. Super learners for time-to-event outcomes have been developed, but the literature is technical and a reader may find it challenging to gather together the full details of how these methods work and can be implemented. In this paper we provide a practical tutorial on super learner methods for time-to-event outcomes. An overview of the general steps involved in the super learner is given, followed by details of three specific implementations for time-to-event outcomes. We cover discrete-time and continuous-time versions of the super learner, as described by Polley and van der Laan (2011), Westling et al. (2023) and Munch and Gerds (2024). We compare the properties of the methods and provide information on how they can be implemented in R. The methods are illustrated using an open access data set and R code is provided.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 13:44:53 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Keogh",
"Ruth H.",
""
],
[
"Diaz-Ordaz",
"Karla",
""
],
[
"van Geloven",
"Nan",
""
],
[
"Gran",
"Jon Michael",
""
],
[
"Tanner",
"Kamaryn T.",
""
]
] |
2509.03316
|
Matja\v{z} Kukar PhD
|
Fatemeh Azad, Zoran Bosni\'c, Matja\v{z} Kukar
|
Meta-Imputation Balanced (MIB): An Ensemble Approach for Handling Missing Data in Biomedical Machine Learning
| null | null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Missing data represents a fundamental challenge in machine learning applications, often reducing model performance and reliability. This problem is particularly acute in fields like bioinformatics and clinical machine learning, where datasets are frequently incomplete due to the nature of both data generation and data collection. While numerous imputation methods exist, from simple statistical techniques to advanced deep learning models, no single method consistently performs well across diverse datasets and missingness mechanisms. This paper proposes a novel Meta-Imputation approach that learns to combine the outputs of multiple base imputers to predict missing values more accurately. By training the proposed method called Meta-Imputation Balanced (MIB) on synthetically masked data with known ground truth, the system learns to predict the most suitable imputed value based on the behavior of each method. Our work highlights the potential of ensemble learning in imputation and paves the way for more robust, modular, and interpretable preprocessing pipelines in real-world machine learning systems.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 13:49:54 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Azad",
"Fatemeh",
""
],
[
"Bosnić",
"Zoran",
""
],
[
"Kukar",
"Matjaž",
""
]
] |
2509.03318
|
Eduard Kamburjan
|
Eduard Kamburjan, Vidar Norstein Klungre, Yuanwei Qu, Rudolf Schlatte, Egor V. Kostylev, Martin Giese, Einar Broch Johnsen
|
Semantically Reflected Programs
| null | null | null | null |
cs.PL cs.LO
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
This paper addresses the dichotomy between the formalization of structural and the formalization of behavioral knowledge by means of semantically lifted programs, which explore an intuitive connection between programs and knowledge graphs. While knowledge graphs and ontologies are eminently useful to represent formal knowledge about a system's individuals and universals, programming languages are designed to describe the system's evolution. To address this dichotomy, we introduce a semantic lifting of the program states of an executing program into a knowledge graph, for an object-oriented programming language. The resulting graph is exposed as a semantic reflection layer within the programming language, allowing programmers to leverage knowledge of the application domain in their programs. In this paper, we formalize semantic lifting and semantic reflection for a small programming language, SMOL, explain the operational aspects of the language, and consider type correctness and virtualisation for runtime program queries through the semantic reflection layer. We illustrate semantic lifting and semantic reflection through a case study of geological modelling and discuss different applications of the technique. The language implementation is open source and available online.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 13:52:50 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Kamburjan",
"Eduard",
""
],
[
"Klungre",
"Vidar Norstein",
""
],
[
"Qu",
"Yuanwei",
""
],
[
"Schlatte",
"Rudolf",
""
],
[
"Kostylev",
"Egor V.",
""
],
[
"Giese",
"Martin",
""
],
[
"Johnsen",
"Einar Broch",
""
]
] |
2509.03319
|
Joel Jaskari
|
Joel Jaskari, Chandreyee Roy, Fumiko Ogushi, Mikko Saukkoriipi, Jaakko Sahlsten, Kimmo Kaski
|
Temporal social network modeling of mobile connectivity data with graph neural networks
|
22 pages, 7 figures
| null | null | null |
cs.SI cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Graph neural networks (GNNs) have emerged as a state-of-the-art data-driven tool for modeling connectivity data of graph-structured complex networks and integrating information of their nodes and edges in space and time. However, as of yet, the analysis of social networks using the time series of people's mobile connectivity data has not been extensively investigated. In the present study, we investigate four snapshot - based temporal GNNs in predicting the phone call and SMS activity between users of a mobile communication network. In addition, we develop a simple non - GNN baseline model using recently proposed EdgeBank method. Our analysis shows that the ROLAND temporal GNN outperforms the baseline model in most cases, whereas the other three GNNs perform on average worse than the baseline. The results show that GNN based approaches hold promise in the analysis of temporal social networks through mobile connectivity data. However, due to the relatively small performance margin between ROLAND and the baseline model, further research is required on specialized GNN architectures for temporal social network analysis.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 13:53:03 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Jaskari",
"Joel",
""
],
[
"Roy",
"Chandreyee",
""
],
[
"Ogushi",
"Fumiko",
""
],
[
"Saukkoriipi",
"Mikko",
""
],
[
"Sahlsten",
"Jaakko",
""
],
[
"Kaski",
"Kimmo",
""
]
] |
2509.03320
|
Partha Goswami
|
Partha Goswami
|
Investigation of non-Hermitian and Hermitian models of Altermagnets
|
32 pages, 6 figures. Comments are welcome
| null | null | null |
cond-mat.mes-hall cond-mat.mtrl-sci
|
http://creativecommons.org/licenses/by/4.0/
|
Insulating altermagnets like MnTe exhibit spin configurations where opposing spins are not only aligned antiparallel but also rotated relative to each other. This is an arrangement reminiscent of antiferromagnetism with a twist of spin canting. This study investigates a model Hamiltonian that captures the essential physics of such systems, incorporating key interactions including Dzyaloshinskii-Moriya and conventional exchange terms, relativistic spin-orbit coupling, and d-wave and g-wave orderings. Non-Hermitian dynamics are introduced through complex potentials that simulate energy dissipation and amplification. The paper delves into the behavior of the quantum geometric tensor and the emergence of the quantum anomalous Hall effect within the topologically insulating regime. It also broadens the scope to encompass non-Hermitian metallic altermagnets, focusing on phases characterized by symmetry-breaking d-wave and g-wave order parameters.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 13:53:28 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Goswami",
"Partha",
""
]
] |
2509.03321
|
Linyu Ou
|
Linyu Ou
|
Empowering Lightweight MLLMs with Reasoning via Long CoT SFT
| null | null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
While Reinforcement Learning with Verifiable Rewards has enhanced the reasoning of large-scale language models (LLMs), its efficacy for lightweight multimodal language models (MLLMs) with fewer than seven billion parameters remains underexplored. This paper investigates the role of long Chain-of-Thought (long CoT) data in enhancing the reasoning abilities of such MLLMs. Our findings demonstrate that Supervised Fine-Tuning (SFT) with long CoT data significantly improves MLLM reasoning. Furthermore, we observe that after this initial SFT phase, MLLMs can achieve additional performance gains through a subsequent RL stage. We conclude that a SFT stage with long CoT data is a critical prerequisite for developing the reasoning capabilities of lightweight MLLMs.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 13:53:29 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Ou",
"Linyu",
""
]
] |
2509.03322
|
Yi Zhang
|
Yi Zhang, Dang Liu, Qiaoyan Yu, Ruijun Xi, Xingsen Chen, Shasha Xue, Jice Sun, Xian Du, Xuhui Ning, Tingwen Miao, Pengyu Hu, Hao Yang, Dandan Guan, Xiaoxue Liu, Liang Liu, Yaoyi Li, Shiyong Wang, Canhua Liu, Haijiao Ji, Noah F. Q. Yuan, Hao Zheng, and Jinfeng Jia
|
Fabrication and Characterization of the Moir\'e surface state on a topological insulator
|
16 pages,4 figures,1 table
|
Nano Letter 2025,25,35,13341,13346
|
10.1021/acs.nanolett.5c03535
| null |
cond-mat.mtrl-sci
|
http://creativecommons.org/licenses/by/4.0/
|
A Moire superlattice on the topological insulator surface is predicted to exhibit many novel properties but has not been experimentally realized. Here, we developed a two-step growth method to successfully fabricate a topological insulator Sb2Te3 thin film with a Moire superlattice, which is generated by a twist of the topmost layer via molecular beam epitaxy. The established Moire topological surface state is characterized by scanning tunneling microscopy and spectroscopy. By application of a magnetic field, new features in Landau levels arise on the Moire region compared to the pristine surface of Sb2Te3, which makes the system a promising platform for pursuing next-generation electronics. Notably, the growth method, which circumvents contamination and the induced interface defects in the manual fabrication method, can be widely applied to other van der Waals materials for fabricating Moire superlattices.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 13:57:59 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Zhang",
"Yi",
""
],
[
"Liu",
"Dang",
""
],
[
"Yu",
"Qiaoyan",
""
],
[
"Xi",
"Ruijun",
""
],
[
"Chen",
"Xingsen",
""
],
[
"Xue",
"Shasha",
""
],
[
"Sun",
"Jice",
""
],
[
"Du",
"Xian",
""
],
[
"Ning",
"Xuhui",
""
],
[
"Miao",
"Tingwen",
""
],
[
"Hu",
"Pengyu",
""
],
[
"Yang",
"Hao",
""
],
[
"Guan",
"Dandan",
""
],
[
"Liu",
"Xiaoxue",
""
],
[
"Liu",
"Liang",
""
],
[
"Li",
"Yaoyi",
""
],
[
"Wang",
"Shiyong",
""
],
[
"Liu",
"Canhua",
""
],
[
"Ji",
"Haijiao",
""
],
[
"Yuan",
"Noah F. Q.",
""
],
[
"Zheng",
"Hao",
""
],
[
"Jia",
"Jinfeng",
""
]
] |
2509.03323
|
Xizhe Zhang
|
Xizhe Zhang and Jiayang Zhu
|
Heatmap Guided Query Transformers for Robust Astrocyte Detection across Immunostains and Resolutions
| null | null | null | null |
cs.CV cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Astrocytes are critical glial cells whose altered morphology and density are hallmarks of many neurological disorders. However, their intricate branching and stain dependent variability make automated detection of histological images a highly challenging task. To address these challenges, we propose a hybrid CNN Transformer detector that combines local feature extraction with global contextual reasoning. A heatmap guided query mechanism generates spatially grounded anchors for small and faint astrocytes, while a lightweight Transformer module improves discrimination in dense clusters. Evaluated on ALDH1L1 and GFAP stained astrocyte datasets, the model consistently outperformed Faster R-CNN, YOLOv11 and DETR, achieving higher sensitivity with fewer false positives, as confirmed by FROC analysis. These results highlight the potential of hybrid CNN Transformer architectures for robust astrocyte detection and provide a foundation for advanced computational pathology tools.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:01:04 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Zhang",
"Xizhe",
""
],
[
"Zhu",
"Jiayang",
""
]
] |
2509.03324
|
Yixiong Jing
|
Yixiong Jing, Cheng Zhang, Haibing Wu, Guangming Wang, Olaf Wysocki, Brian Sheil
|
InfraDiffusion: zero-shot depth map restoration with diffusion models and prompted segmentation from sparse infrastructure point clouds
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Point clouds are widely used for infrastructure monitoring by providing geometric information, where segmentation is required for downstream tasks such as defect detection. Existing research has automated semantic segmentation of structural components, while brick-level segmentation (identifying defects such as spalling and mortar loss) has been primarily conducted from RGB images. However, acquiring high-resolution images is impractical in low-light environments like masonry tunnels. Point clouds, though robust to dim lighting, are typically unstructured, sparse, and noisy, limiting fine-grained segmentation. We present InfraDiffusion, a zero-shot framework that projects masonry point clouds into depth maps using virtual cameras and restores them by adapting the Denoising Diffusion Null-space Model (DDNM). Without task-specific training, InfraDiffusion enhances visual clarity and geometric consistency of depth maps. Experiments on masonry bridge and tunnel point cloud datasets show significant improvements in brick-level segmentation using the Segment Anything Model (SAM), underscoring its potential for automated inspection of masonry assets. Our code and data is available at https://github.com/Jingyixiong/InfraDiffusion-official-implement.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:01:10 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Jing",
"Yixiong",
""
],
[
"Zhang",
"Cheng",
""
],
[
"Wu",
"Haibing",
""
],
[
"Wang",
"Guangming",
""
],
[
"Wysocki",
"Olaf",
""
],
[
"Sheil",
"Brian",
""
]
] |
2509.03325
|
Sophia Stasevic
|
S. Stasevic, J. Milli, J. Mazoyer, A.-M. Lagrange, S. Bergeon
|
Optimising reference library selection for reference-star differential imaging of discs with SPHERE/IRDIS
|
Main article: 11 pages, 5 figures, 2 tables; Appendix: 5 pages, 7 figures, 1 table
| null |
10.1051/0004-6361/202555338
| null |
astro-ph.EP astro-ph.IM astro-ph.SR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The direct detection of circumstellar discs through high-contrast imaging provides key insights into the history and dynamics of planetary systems. Pole-on discs, especially faint debris discs, are difficult to detect and require careful consideration during post-processing to remove stellar residuals from the data while preserving the disc signal. Reference-star differential imaging (RDI) serves as one of the primary post-processing methods for disc observations. We aim to develop a method of reference frame selection that is optimised for the reduction of pole-on discs. Method: We performed principal component analysis based RDI on seven known disc targets and 20 disc-free targets with varying observational conditions, using reference libraries built from frames preselected to best match different observational, atmospheric, and stellar parameters of the science frames. The contrast of the disc-free reductions was measured, and forward modelling was used to estimate the signal loss from over-subtraction using synthetic pole-on discs with two different widths and four different radii. The signal-to-noise ratio (S/N) of the real disc targets was measured. Results: Diverse reference libraries built using subsets of frames that closely matched different parameters achieved the best disc S/N and smallest deviation from the best contrast of each target, outperforming libraries built using a single criterion as a selection metric. Libraries built using frame-to-frame Pearson correlation coefficient alone as a selection criterion achieved the best mean contrast overall. Both selection metrics performed consistently well for all disc radii and observational conditions. We also found that reference libraries built using frames observed close in time to the science frame performed well for discs at small separations, giving the best contrast for ~30% of the targets at a radius of 20px.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:01:40 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Stasevic",
"S.",
""
],
[
"Milli",
"J.",
""
],
[
"Mazoyer",
"J.",
""
],
[
"Lagrange",
"A. -M.",
""
],
[
"Bergeon",
"S.",
""
]
] |
2509.03326
|
Dmitry Shestakov
|
D. A. Shestakov and V. V. Izmodenov
|
Numerical Modeling of Galactic Cosmic Ray Modulation in the Heliosphere
|
Submitted to Lobachevskii Journal of Mathematics
| null | null | null |
astro-ph.HE astro-ph.IM astro-ph.SR physics.comp-ph physics.plasm-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Numerical modeling of galactic cosmic rays (GCRs) penetration through the heliosphere to the vicinity of the Sun is considered. Galactic cosmic rays are charged particles with energies exceeding 10 MeV/nucl., originating from far beyond the boundaries of our Solar System. As they penetrate through the heliosphere - the region of space filled by the solar wind - they interact strongly with the interplanetary magnetic field. In this paper, we present numerical approaches to solving the so-called Parker transport equation for the isotropic velocity distribution function of GCRs. This equation includes a convective term, anisotropic diffusion, adiabatic cooling, and drifts. Additionally, the diffusion coefficient is spatially and energy-dependent, varying by several orders of magnitude. Our numerical approaches are based on the finite-difference method (Crank-Nicolson scheme) and the stochastic differential equations (SDE) method. The numerical methods were validated against a known analytical solution under simplified conditions. For the general problem formulation, which involves anisotropic diffusion and the Parker spiral interplanetary magnetic field configuration, we used the most efficient and flexible SDE method and compared the numerical results with the data from the works of Kota & Jokipii (1983) and Burger (2012). Special attention was devoted to incorporating drift along the heliospheric current sheet in the model.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:02:09 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Shestakov",
"D. A.",
""
],
[
"Izmodenov",
"V. V.",
""
]
] |
2509.03327
|
Wenxin Li
|
Wenxin Li, Jonathan T. Reichanadter, Shan Wu, Ji Seop Oh, Rourav Basak, Shannon C. Haley, Elio Vescovo, Donghui Lu, Makoto Hashimoto, Christoph Klewe, Suchismita Sarker, James G. Analytis, Robert J. Birgeneau, Jeffrey B. Neaton, Yu He
|
Role of Fe intercalation on the electronic correlation in resistively switchable antiferromagnet Fe$_{x}$NbS$_2$
| null | null | null | null |
cond-mat.str-el
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Among the family of intercalated transition-metal dichalcogenides (TMDs), Fe$_{x}$NbS$_2$ is found to possess unique current-induced resistive switching behaviors, tunable antiferromagnetic states, and a commensurate charge order, all of which are tied to a critical Fe doping of $x_c$ = 1/3. However, the electronic origin of such extreme stoichiometry sensitivities remains unclear. Combining angle-resolved photoemission spectroscopy (ARPES) with density functional theory (DFT) calculations, we identify and characterize a dramatic eV-scale electronic restructuring that occurs across the $x_c$. Moment-carrying Fe 3$d_{z^2}$ electrons manifest as narrow bands within 200 meV to the Fermi level, distinct from other transition metal intercalated TMD magnets. This state strongly interacts with the itinerant electron in TMD layer, and rapidly loses coherence above $x_c$. These observations resemble the exceptional electronic and magnetic sensitivity of strongly correlated systems upon charge doping, shedding light on the important role of electronic correlation in magnetic TMDs.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:04:01 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Li",
"Wenxin",
""
],
[
"Reichanadter",
"Jonathan T.",
""
],
[
"Wu",
"Shan",
""
],
[
"Oh",
"Ji Seop",
""
],
[
"Basak",
"Rourav",
""
],
[
"Haley",
"Shannon C.",
""
],
[
"Vescovo",
"Elio",
""
],
[
"Lu",
"Donghui",
""
],
[
"Hashimoto",
"Makoto",
""
],
[
"Klewe",
"Christoph",
""
],
[
"Sarker",
"Suchismita",
""
],
[
"Analytis",
"James G.",
""
],
[
"Birgeneau",
"Robert J.",
""
],
[
"Neaton",
"Jeffrey B.",
""
],
[
"He",
"Yu",
""
]
] |
2509.03328
|
Pierre Faug\`ere
|
Pierre Faug\`ere, Cyril Labb\'e
|
Dynamical interface above a hard wall and reflected SPDE on the half-line
| null | null | null | null |
math.PR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We consider a dynamical random interface on the infinite lattice $\mathbb{N}$ evolving according to a "corner flip" dynamic above a hard wall, with an additional pinning at the origin. We study the stationary fluctuations under a diffusive scaling and prove convergence in law towards the solution of an SPDE of Nualart-Pardoux's type, namely the Reflected Stochastic Heat Equation on the half-line. We also obtain that the law of the 3-dimensional Bessel process is an invariant measure for this SPDE.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:04:49 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Faugère",
"Pierre",
""
],
[
"Labbé",
"Cyril",
""
]
] |
2509.03329
|
Mateo Dulce Rubio
|
Melissa Robles, Catalina Bernal, Denniss Raigoso, Mateo Dulce Rubio
|
SESGO: Spanish Evaluation of Stereotypical Generative Outputs
| null | null | null | null |
cs.CY cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
This paper addresses the critical gap in evaluating bias in multilingual Large Language Models (LLMs), with a specific focus on Spanish language within culturally-aware Latin American contexts. Despite widespread global deployment, current evaluations remain predominantly US-English-centric, leaving potential harms in other linguistic and cultural contexts largely underexamined. We introduce a novel, culturally-grounded framework for detecting social biases in instruction-tuned LLMs. Our approach adapts the underspecified question methodology from the BBQ dataset by incorporating culturally-specific expressions and sayings that encode regional stereotypes across four social categories: gender, race, socioeconomic class, and national origin. Using more than 4,000 prompts, we propose a new metric that combines accuracy with the direction of error to effectively balance model performance and bias alignment in both ambiguous and disambiguated contexts. To our knowledge, our work presents the first systematic evaluation examining how leading commercial LLMs respond to culturally specific bias in the Spanish language, revealing varying patterns of bias manifestation across state-of-the-art models. We also contribute evidence that bias mitigation techniques optimized for English do not effectively transfer to Spanish tasks, and that bias patterns remain largely consistent across different sampling temperatures. Our modular framework offers a natural extension to new stereotypes, bias categories, or languages and cultural contexts, representing a significant step toward more equitable and culturally-aware evaluation of AI systems in the diverse linguistic environments where they operate.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:04:51 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Robles",
"Melissa",
""
],
[
"Bernal",
"Catalina",
""
],
[
"Raigoso",
"Denniss",
""
],
[
"Rubio",
"Mateo Dulce",
""
]
] |
2509.03330
|
Bnaya Gross
|
Bnaya Gross, Joseph Ehlert, Vadim N. Gladyshev, Joseph Loscalzo, and Albert-L\'aszl\'o Barab\'asi
|
Network-driven discovery of repurposable drugs targeting hallmarks of aging
| null | null | null | null |
q-bio.MN physics.soc-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Despite the thousands of genes implicated in age-related phenotypes, effective interventions for aging remain elusive, a lack of advance rooted in the multifactorial nature of longevity and the functional interconnectedness of the molecular components implicated in aging. Here, we introduce a network medicine framework that integrates 2,358 longevity-associated genes onto the human interactome to identify existing drugs that can modulate aging processes. We find that genes associated with each hallmark of aging form a connected subgraph, or hallmark module, a discovery enabling us to measure the proximity of 6,442 clinically approved or experimental compounds to each hallmark. We then introduce a transcription-based metric, $pAGE$, which evaluates whether the drug-induced expression shifts reinforce or counteract known age-related expression changes. By integrating network proximity and $pAGE$, we identify multiple drug repurposing candidate that not only target specific hallmarks but act to reverse their aging-associated transcriptional changes. Our findings are interpretable, revealing for each drug the molecular mechanisms through which it modulates the hallmark, offering an experimentally falsifiable framework to leverage genomic discoveries to accelerate drug repurposing for longevity.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:05:56 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Gross",
"Bnaya",
""
],
[
"Ehlert",
"Joseph",
""
],
[
"Gladyshev",
"Vadim N.",
""
],
[
"Loscalzo",
"Joseph",
""
],
[
"Barabási",
"Albert-László",
""
]
] |
2509.03331
|
Wei Ma
|
Weizhe Wang, Wei Ma, Qiang Hu, Yao Zhang, Jianfei Sun, Bin Wu, Yang Liu, Guangquan Xu, and Lingxiao Jiang
|
VulnRepairEval: An Exploit-Based Evaluation Framework for Assessing Large Language Model Vulnerability Repair Capabilities
| null | null | null | null |
cs.SE cs.CR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The adoption of Large Language Models (LLMs) for automated software vulnerability patching has shown promising outcomes on carefully curated evaluation sets. Nevertheless, existing datasets predominantly rely on superficial validation methods rather than exploit-based verification, leading to overestimated performance in security-sensitive applications. This paper introduces VulnRepairEval, an evaluation framework anchored in functional Proof-of-Concept (PoC) exploits. Our framework delivers a comprehensive, containerized evaluation pipeline that enables reproducible differential assessment, where repair success requires the original exploit to fail execution against the modified code. The benchmark construction involved extensive data curation: we processed over 400 CVEs and approximately 2,500 potential sources to extract a collection of authentic vulnerability instances (23 Python CVEs) amenable to automated testing with working PoCs. Through VulnRepairEval, we conduct a comprehensive evaluation of 12 popular LLMs and observe a significant performance deficit: even the top-performing model successfully addresses merely 5/23 instances (about 21.7%), exposing critical weaknesses in security-focused applications. Our failure analysis reveals that most unsuccessful attempts stem from imprecise vulnerability identification and patches containing syntactic or semantic errors. Enhanced prompting strategies and multi-agent approaches yield minimal improvements, with overall effectiveness remaining largely unaffected. This work contributes a stringent, practical evaluation framework for LLM-driven vulnerability remediation and underscores the necessity for assessment protocols that authentically reflect real-world exploitation scenarios.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:06:10 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Wang",
"Weizhe",
""
],
[
"Ma",
"Wei",
""
],
[
"Hu",
"Qiang",
""
],
[
"Zhang",
"Yao",
""
],
[
"Sun",
"Jianfei",
""
],
[
"Wu",
"Bin",
""
],
[
"Liu",
"Yang",
""
],
[
"Xu",
"Guangquan",
""
],
[
"Jiang",
"Lingxiao",
""
]
] |
2509.03332
|
Qing Yang
|
Qing Yang, Gu-yue Zhang, Yi Huang, and Xiao Guo
|
PTA Frequency Band Individual Gravitational Wave Sources and Dark Energy Detection Based on Cosmological Simulation
|
15 pages, 8 figures, 5 tables
| null | null | null |
astro-ph.CO astro-ph.HE gr-qc
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Nanohertz gravitational waves (GWs) from supermassive binary black holes (SMBBHs), detectable via pulsar timing arrays (PTAs), offer a novel avenue to constrain dark energy. Based on cosmological simulations and semi-analytic galaxy formation models, this study explores the detectability of individual nanohertz SMBBH sources using next-generation PTAs and their potential for constraining dark energy under an optimistic scenario considering only the presence of white noise. By constructing light-cone SMBBH populations across hardening timescales ($\tau_H = 0.1/5/10$Gyr) and computing signal-to-noise ratios (SNR), we find advanced PTAs can resolve $10^2$--$10^3$ sources with SNR $> 8$ (primarily at $z < 1$ with chirp masses of $10^8$--$10^{10}M_{\odot}$). If electromagnetic counterparts can be identified, optimal configurations ($\sigma_t = 50$ns, $N_p = 1000$, $T_{\text{obs}} = 30$yr with$ \tau_H \leq 5$Gyr) could constrain the dark energy equation-of-state (EoS) parameter $w$ to $\Delta w \sim 0.023$--$0.048$, where the constraints only exhibit weak dependence on $\tau_H$ within $0.1$--$5$Gyr. If only $10\%$ of GW sources have detectable electromagnetic counterparts, constraints weaken to $\Delta w = 0.075$ ($\tau_H = 0.1$Gyr) and $\Delta w = 0.162$ ($\tau_H = 5$Gyr) under the most optimal parameter configuration. What's more, conservative PTAs ($N_p = 500$, $\sigma_t = 100$--$200$ns) with additional $30$-year data accumulation could double resolvable source counts and improve $\Delta w$ precision by $\sim 40\%$.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:07:56 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Yang",
"Qing",
""
],
[
"Zhang",
"Gu-yue",
""
],
[
"Huang",
"Yi",
""
],
[
"Guo",
"Xiao",
""
]
] |
2509.03333
|
Tianfu Qi
|
Tianfu Qi, Jun Wang
|
Baseband Model, Cutoff Rate Bounds and Constellation Shaping for Mixed Gaussian-Impulsive Noise
| null | null | null | null |
eess.SP cs.IT math.IT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Mixed noise, composed of white Gaussian noise (WGN) and impulsive noise (IN), appears in numerous communication scenarios and can severely degrade system performance. In this paper, we address this issue by optimizing the transmitted constellation under mixed noise based on a theoretical analysis of the cutoff rate (CR). First, starting from the passband model of the mixed noise, we derive its corresponding baseband representation. Due to the complexity of the CR, an exact analytic expression is generally intractable. Therefore, the baseband noise model is employed to obtain closed-form lower and upper bounds of the CR. A piecewise linear approximation is applied to derive efficient bounds by exploiting the algebraic properties of the integral terms. These bounds are then used as criteria to optimize the transmitted constellation points in both geometric and probabilistic distributions. The projected gradient method is employed to solve the optimization problem, and the convergence and properties of the solutions are analyzed. Numerical results demonstrate that the proposed CR bounds are tight and exhibit the expected asymptotic behavior. Furthermore, the optimized constellation scheme achieves a significant rate improvement compared to baselines.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:09:44 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Qi",
"Tianfu",
""
],
[
"Wang",
"Jun",
""
]
] |
2509.03334
|
Roshan Trivedi
|
Roshan S. Trivedi, Advik D. Vira, Brant M. Jones, Katherine D. Burgess, Ziyu Huang, Honglin Liu, Pranav Rane, Mengkun Tian, Masatoshi Hirabayashi, Thomas M. Orlando, Zhigang Jiang, Phillip N. First
|
Creation of Lunar-Like Rims in Ilmenite using Synthetic Solar Wind
|
26 pages, 13 figures, includes Supplementary Information
| null | null | null |
astro-ph.EP
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Space weathering of lunar minerals, due to bombardment from solar wind (SW) particles and micrometeoroid impacts, modifies the mineralogy within tens of nanometers of the surface, i.e., the rim. Via remote sensing, spectroscopic signatures of these modifications have long been used to gauge surface exposure times on the Moon. However, the relative contributions of the solar wind and micrometeroids in the creation of rim features are still debated, particularly for the few-nm clusters known as nanophase iron (npFe$^0$), which commonly form in ferrous minerals. We address this issue in the laboratory, using deuterium ions and low-energy electrons as a synthetic solar wind plasma to irradiate ilmenite FeTiO$_3$), a common lunar mineral. Characterization by high-resolution scanning transmission electron microscopy and electron energy-loss spectroscopy shows that the solar wind alone creates rims with all the main characteristics of lunar samples. We conclusively identify npFe$^0$ and quantify its distribution as a function of depth and fluence, allowing us to estimate the SW exposure of Apollo soil 71501. Our results confirm that small npFe$^0$ (<10 nm in diameter) form due to the solar wind. Similar experiments could provide microscopic details of space weathering, improving the link between surface modification processes and macroscopic remote sensing data.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:10:11 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Trivedi",
"Roshan S.",
""
],
[
"Vira",
"Advik D.",
""
],
[
"Jones",
"Brant M.",
""
],
[
"Burgess",
"Katherine D.",
""
],
[
"Huang",
"Ziyu",
""
],
[
"Liu",
"Honglin",
""
],
[
"Rane",
"Pranav",
""
],
[
"Tian",
"Mengkun",
""
],
[
"Hirabayashi",
"Masatoshi",
""
],
[
"Orlando",
"Thomas M.",
""
],
[
"Jiang",
"Zhigang",
""
],
[
"First",
"Phillip N.",
""
]
] |
2509.03336
|
Balu Bhasuran
|
Sharanya Manoharan, Balu Bhasuran, Oviya Ramalakshmi Iyyappan, Mohamed Saleem Abdul Shukkoor, Malathi Sellapan, Kalpana Raja
|
AI-Driven Drug Repurposing through miRNA-mRNA Relation
| null | null | null | null |
q-bio.MN cs.IR q-bio.QM
|
http://creativecommons.org/licenses/by/4.0/
|
miRNA mRNA relations are closely linked to several biological processes and disease mechanisms In a recent study we tested the performance of large language models LLMs on extracting miRNA mRNA relations from PubMed PubMedBERT achieved the best performance of 0.783 F1 score for miRNA mRNA Interaction Corpus MMIC Here we first applied the finetuned PubMedBERT model to extract miRNA mRNA relations from PubMed for chronic obstructive pulmonary disease COPD Alzheimers disease AD stroke type 2 diabetes mellitus T2DM chronic liver disease and cancer Next we retrieved miRNA drug relations using KinderMiner a literature mining tool for relation extraction Then we constructed three interaction networks 1 disease centric network 2 drug centric network and 3 miRNA centric network comprising 3497 nodes and 16417 edges organized as a directed graph to capture complex biological relationships Finally we validated the drugs using MIMIC IV Our integrative approach revealed both established and novel candidate drugs for diseases under study through 595 miRNA drug relations extracted from PubMed To the best of our knowledge this is the first study to systematically extract and visualize relationships among four distinct biomedical entities miRNA mRNA drug and disease
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:12:32 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Manoharan",
"Sharanya",
""
],
[
"Bhasuran",
"Balu",
""
],
[
"Iyyappan",
"Oviya Ramalakshmi",
""
],
[
"Shukkoor",
"Mohamed Saleem Abdul",
""
],
[
"Sellapan",
"Malathi",
""
],
[
"Raja",
"Kalpana",
""
]
] |
2509.03337
|
Guanghui Zhang
|
Liren Lin, Guanghui Zhang, Bocong Chen and Hongwei Liu
|
New Bounds for Linear Codes with Applications
|
15 pages
| null | null | null |
cs.IT math.IT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Bounds on linear codes play a central role in coding theory, as they capture the fundamental trade-off between error-correction capability (minimum distance) and information rate (dimension relative to length). Classical results characterize this trade-off solely in terms of the parameters $n$, $k$, $d$ and $q$. In this work we derive new bounds under the additional assumption that the code contains a nonzero codeword of weight $w$.By combining residual-code techniques with classical results such as the Singleton and Griesmer bounds,we obtain explicit inequalities linking $n$, $k$, $d$, $q$ and $w$. These bounds impose sharper restrictions on admissible codeword weights, particularly those close to the minimum distance or to the code length. Applications include refined constraints on the weights of MDS codes, numerical restrictions on general linear codes, and excluded weight ranges in the weight distribution. Numerical comparisons across standard parameter sets demonstrate that these $w$-aware bounds strictly enlarge known excluded weight ranges and sharpen structural limitations on linear codes.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:14:13 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Lin",
"Liren",
""
],
[
"Zhang",
"Guanghui",
""
],
[
"Chen",
"Bocong",
""
],
[
"Liu",
"Hongwei",
""
]
] |
2509.03339
|
K. V. Krishna
|
Khyodeno Mozhui and Tithi Dwary and K. V. Krishna
|
Line Graphs of Non-Word-Representable Graphs are Not Always Non-Word-Representable
| null | null | null | null |
math.CO cs.DM
|
http://creativecommons.org/licenses/by/4.0/
|
A graph is said to be word-representable if there exists a word over its vertex set such that any two vertices are adjacent if and only if they alternate in the word. If no such word exists, the graph is non-word-representable. In the literature, there are examples of non-word-representable graphs whose line graphs are non-word-representable. However, it is an open problem to determine whether the line graph of a non-word-representable graph is always non-word-representable or not? In this work, we address the open problem by considering a class of non-word-representable graphs, viz., Mycielski graphs of odd cycles of length at least five, and show that their line graphs are word-representable.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:17:05 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Mozhui",
"Khyodeno",
""
],
[
"Dwary",
"Tithi",
""
],
[
"Krishna",
"K. V.",
""
]
] |
2509.03340
|
Fleur Hendriks
|
Fleur Hendriks, Ond\v{r}ej Roko\v{s}, Martin Do\v{s}k\'a\v{r}, Marc G.D. Geers, Vlado Menkovski
|
Equivariant Flow Matching for Symmetry-Breaking Bifurcation Problems
|
12 pages, 7 figures including appendices
| null | null | null |
cs.LG cs.AI cs.CE physics.comp-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Bifurcation phenomena in nonlinear dynamical systems often lead to multiple coexisting stable solutions, particularly in the presence of symmetry breaking. Deterministic machine learning models struggle to capture this multiplicity, averaging over solutions and failing to represent lower-symmetry outcomes. In this work, we propose a generative framework based on flow matching to model the full probability distribution over bifurcation outcomes. Our method enables direct sampling of multiple valid solutions while preserving system symmetries through equivariant modeling. We introduce a symmetric matching strategy that aligns predicted and target outputs under group actions, allowing accurate learning in equivariant settings. We validate our approach on a range of systems, from toy models to complex physical problems such as buckling beams and the Allen-Cahn equation. Our results demonstrate that flow matching significantly outperforms non-probabilistic and variational methods in capturing multimodal distributions and symmetry-breaking bifurcations, offering a principled and scalable solution for modeling multistability in high-dimensional systems.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:18:05 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Hendriks",
"Fleur",
""
],
[
"Rokoš",
"Ondřej",
""
],
[
"Doškář",
"Martin",
""
],
[
"Geers",
"Marc G. D.",
""
],
[
"Menkovski",
"Vlado",
""
]
] |
2509.03341
|
Ilana Sebag
|
Ilana Sebag, Jean-Yves Franceschi, Alain Rakotomamonjy, Alexandre Allauzen, Jamal Atif
|
On the MIA Vulnerability Gap Between Private GANs and Diffusion Models
| null | null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Generative Adversarial Networks (GANs) and diffusion models have emerged as leading approaches for high-quality image synthesis. While both can be trained under differential privacy (DP) to protect sensitive data, their sensitivity to membership inference attacks (MIAs), a key threat to data confidentiality, remains poorly understood. In this work, we present the first unified theoretical and empirical analysis of the privacy risks faced by differentially private generative models. We begin by showing, through a stability-based analysis, that GANs exhibit fundamentally lower sensitivity to data perturbations than diffusion models, suggesting a structural advantage in resisting MIAs. We then validate this insight with a comprehensive empirical study using a standardized MIA pipeline to evaluate privacy leakage across datasets and privacy budgets. Our results consistently reveal a marked privacy robustness gap in favor of GANs, even in strong DP regimes, highlighting that model type alone can critically shape privacy leakage.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:18:22 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Sebag",
"Ilana",
""
],
[
"Franceschi",
"Jean-Yves",
""
],
[
"Rakotomamonjy",
"Alain",
""
],
[
"Allauzen",
"Alexandre",
""
],
[
"Atif",
"Jamal",
""
]
] |
2509.03342
|
Ramon Castaneda-Priego Dr.
|
Gabriel P\'erez-Angel, Marco A. Ram\'irez-Gu\'izar, N\'estor M. De Los Santos-L\'opez, Jos\'e M. M\'endez-Alcaraz, Ram\'on Casta\~neda-Priego
|
Quantifying many-body contributions to depletion forces
| null | null | null | null |
cond-mat.soft
|
http://creativecommons.org/licenses/by/4.0/
|
Effective interactions inherently encompass many-body effects that appear unified. Analyzing these in reverse, that is, separating them into contributions from pairs, triples, or larger groups, is typically intricate and seldom pursued. However, this could offer essential insights into the structural architecture of complex systems, such as soft materials. This contribution tackles this issue employing a new simulation-based approach to accurately determine effective interactions. The approach is sufficiently sensitive to assess the effects of higher-order terms in the depletion potentials between large particles. Previous research has primarily focused on the role of small particles, and this work expands on these findings. However, understanding the contributions of large particles on the effective forces, particularly beyond the dilute limit, remains challenging and is not yet fully grasped, leading us to primarily focus on exploring this topic. Within the range of particle concentrations examined here, while maintaining a constant chemical potential for smaller particles, we have observed that the concentration of larger particles has no impact on the entropic potential between large colloids, as long as the size ratio remains below $q=0.15$. This confirms a long-established prediction founded on purely geometric considerations, which we have confirmed by means of direct observation. In contrast, for mixtures with less size asymmetry, such effects become considerably influential. Specifically, we have conducted an in-depth examination of scenarios with size asymmetries of $q=0.45$ and $0.60$. Our results for the depletion forces were directly compared with those obtained from the integral equation theory, which has also enabled us to improve and refine the approaches involved.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:20:02 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Pérez-Angel",
"Gabriel",
""
],
[
"Ramírez-Guízar",
"Marco A.",
""
],
[
"Santos-López",
"Néstor M. De Los",
""
],
[
"Méndez-Alcaraz",
"José M.",
""
],
[
"Castañeda-Priego",
"Ramón",
""
]
] |
2509.03343
|
Maxence Baccara
|
Maxence Baccara
|
Functional Limit Theorems for the range of stable random walks
| null | null | null | null |
math.PR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this paper we establish Functional Limit Theorems for the range of random walks in $\mathbb{Z}^d$ that are in the domain of attraction of a non-degenerate $\beta$-stable process in the weakly transient and recurrent regimes. These results complement the fluctuations obtained at fixed time and the functional limit Theorems obtained in the strongly transient regime. The techniques involve original ideas of Le Gall and Rosen for fluctuations and allow to show tightness in some H\"older space, thus also providing sharp regularity results about the limiting processes. The original motivation of this work is the description of functionals appearing in spatial ecology for consumption of resources induced by random motion. We apply our result to estimate the large fluctuations of energy and mortality for a simple prey predator model.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:20:19 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Baccara",
"Maxence",
""
]
] |
2509.03344
|
Pankaj Mishra
|
Sanu Kumar Gangwar, Rajamanickam Ravisankar, Paulsamy Muruganandam, and Pankaj Kumar Mishra
|
Emergence of spin-mixed superstripe phases in spin-orbit coupled spin-1 condensate
|
19 pages, 15 figures
| null | null | null |
cond-mat.quant-gas
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We numerically investigate the ground state phases and quench dynamics of spin-orbit coupled spin-1 Bose-Einstein condensates with ferromagnetic and antiferromagnetic interactions. For finite Rabi coupling, the system exhibits zero-momentum, elongated zero-momentum, and stripe phases, while in the limit $\Omega\rightarrow 0$, the superstripe wave phases emerge. Varying the attractive density-density ($c_0$) and spin-exchange ($c_2$) interactions induces the transition from stripe and superstripe phases to the elongated zero-momentum phase, characterized by miscibility and polarization. The condensate remains miscible in the zero-momentum phase and partially miscible for the elongated, stripe, and superstripe phases. Quenching in Rabi coupling stabilizes the condensate, while spin-orbit coupling quenching leads to fluctuations or the formation of complex patterns, with the stripe phase exhibiting the vanishing of the zeroth spin component over time. Our results may offer valuable insights into engineering exotic quantum phases in ultracold atomic gases.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:20:49 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Gangwar",
"Sanu Kumar",
""
],
[
"Ravisankar",
"Rajamanickam",
""
],
[
"Muruganandam",
"Paulsamy",
""
],
[
"Mishra",
"Pankaj Kumar",
""
]
] |
2509.03345
|
Yunxin Sun
|
Yunxin Sun, Abulhair Saparov
|
Language Models Do Not Follow Occam's Razor: A Benchmark for Inductive and Abductive Reasoning
| null | null | null | null |
cs.AI cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Reasoning is a core capability in artificial intelligence systems, for which large language models (LLMs) have recently shown remarkable progress. However, most work focuses exclusively on deductive reasoning, which is problematic since other types of reasoning are also essential in solving real-world problems, and they are less explored. This work focuses on evaluating LLMs' inductive and abductive reasoning capabilities. We introduce a programmable and synthetic dataset, InAbHyD (pronounced in-a-bid), where each reasoning example consists of an incomplete world model and a set of observations. The task for the intelligent agent is to produce hypotheses to explain observations under the incomplete world model to solve each reasoning example. We propose a new metric to evaluate the quality of hypotheses based on Occam's Razor. We evaluate and analyze some state-of-the-art LLMs. Our analysis shows that LLMs can perform inductive and abductive reasoning in simple scenarios, but struggle with complex world models and producing high-quality hypotheses, even with popular reasoning-enhancing techniques such as in-context learning and RLVR.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:22:42 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Sun",
"Yunxin",
""
],
[
"Saparov",
"Abulhair",
""
]
] |
2509.03346
|
Alessio Caminata
|
Anna Maria Bigatti, Alessio Caminata, Tor Kristian Ellingsen, Evelina Lanteri, Andrea Sanguineti, Irene Villa
|
Solving Polynomial Systems with Gr\"obner Bases: An Introduction to F4 and FGLM
|
are welcome
| null | null | null |
cs.SC math.AC
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
These notes originate from a reading course held by the authors in the spring of 2024 at the Universit\`a di Genova. They provide a hands-on introduction to the F4 and FGLM algorithms. In addition to the notes, we present two implementations of the algorithms: FGLM in CoCoALib and F4 in Sage. These implementations closely follow the structure of the algorithms as described here and are intended to help readers experiment with them in practice, thereby gaining a deeper understanding.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:23:16 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Bigatti",
"Anna Maria",
""
],
[
"Caminata",
"Alessio",
""
],
[
"Ellingsen",
"Tor Kristian",
""
],
[
"Lanteri",
"Evelina",
""
],
[
"Sanguineti",
"Andrea",
""
],
[
"Villa",
"Irene",
""
]
] |
2509.03347
|
Jiahao Wu
|
Jiahao Wu, Xutun Wang, Yuxin Wu, Guihua Zhang, Jiayue Liu, Xin Li
|
Physics-informed machine learning for combustion: A review
| null | null | null | null |
physics.chem-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Physics-informed machine learning (PIML) represents an emerging paradigm that integrates various forms of physical knowledge into machine learning (ML) components, thereby enhancing the physical consistency of ML models compared to purely data-driven paradigms. The field of combustion, characterized by a rich foundation of physical laws and abundant data, is undergoing a transformation due to PIML. This paper aims to provide a comprehensive overview of PIML for combustion, systematically outlining fundamental principles, significant contributions, key advancements, and available resources. The application of PIML in combustion is categorized into three domains: combustion chemical kinetics, combustion reacting flows, and other combustion-related problems. Additionally, current challenges, potential solutions, and practical guidelines for researchers and engineers will be discussed. A primary focus of this review is to demonstrate how combustion laws can be integrated into ML, either through soft or hard constraints, via loss functions or representation models, and within coordinate-to-variable or field-to-field paradigms. This paper shows that PIML offers a unified framework linking physics, model, and data in combustion--integrating physical knowledge in model-to-data simulation and reconstruction tasks, as well as data-to-model modeling tasks--resulting in enhanced data, improved physical models, and more reliable ML models. PIML for combustion presents significant opportunities for both the combustion and ML communities, encouraging greater collaboration and cross-disciplinary engagement.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:23:28 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Wu",
"Jiahao",
""
],
[
"Wang",
"Xutun",
""
],
[
"Wu",
"Yuxin",
""
],
[
"Zhang",
"Guihua",
""
],
[
"Liu",
"Jiayue",
""
],
[
"Li",
"Xin",
""
]
] |
2509.03348
|
Yewen Li
|
Yewen Li, Jingtong Gao, Nan Jiang, Shuai Mao, Ruyi An, Fei Pan, Xiangyu Zhao, Bo An, Qingpeng Cai, Peng Jiang
|
Generative Auto-Bidding in Large-Scale Competitive Auctions via Diffusion Completer-Aligner
| null | null | null | null |
cs.GT cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Auto-bidding is central to computational advertising, achieving notable commercial success by optimizing advertisers' bids within economic constraints. Recently, large generative models show potential to revolutionize auto-bidding by generating bids that could flexibly adapt to complex, competitive environments. Among them, diffusers stand out for their ability to address sparse-reward challenges by focusing on trajectory-level accumulated rewards, as well as their explainable capability, i.e., planning a future trajectory of states and executing bids accordingly. However, diffusers struggle with generation uncertainty, particularly regarding dynamic legitimacy between adjacent states, which can lead to poor bids and further cause significant loss of ad impression opportunities when competing with other advertisers in a highly competitive auction environment. To address it, we propose a Causal auto-Bidding method based on a Diffusion completer-aligner framework, termed CBD. Firstly, we augment the diffusion training process with an extra random variable t, where the model observes t-length historical sequences with the goal of completing the remaining sequence, thereby enhancing the generated sequences' dynamic legitimacy. Then, we employ a trajectory-level return model to refine the generated trajectories, aligning more closely with advertisers' objectives. Experimental results across diverse settings demonstrate that our approach not only achieves superior performance on large-scale auto-bidding benchmarks, such as a 29.9% improvement in conversion value in the challenging sparse-reward auction setting, but also delivers significant improvements on the Kuaishou online advertising platform, including a 2.0% increase in target cost.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:25:36 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Li",
"Yewen",
""
],
[
"Gao",
"Jingtong",
""
],
[
"Jiang",
"Nan",
""
],
[
"Mao",
"Shuai",
""
],
[
"An",
"Ruyi",
""
],
[
"Pan",
"Fei",
""
],
[
"Zhao",
"Xiangyu",
""
],
[
"An",
"Bo",
""
],
[
"Cai",
"Qingpeng",
""
],
[
"Jiang",
"Peng",
""
]
] |
2509.03349
|
Giovanni Zotti Mr
|
Giovanni Zotti, Dmitri B. Horoshko, Mikhail I. Kolobov, Yoad Michael, Ziv Gefen, Maria V. Chekhova, Kazuki Hashimoto
|
Low-coherence interferometry with undetected mid-infrared photons in the high-gain regime
|
14 pages, 11 figures
| null | null | null |
physics.optics physics.ins-det
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We develop a high-parametric-gain SU(1,1) interferometer based on an aperiodically poled Potassium Titanyl Phosphate (apKTP) crystal, enabling frequency-domain low-coherence interferometry with undetected mid-infrared photons. The system achieves a signal-to-noise ratio as high as 40 dB and axial resolution of 30 $\mu$m, with a 3 $\mu$m-centered idler beam. By increasing the poling-period range, we also improve the axial resolution to 17 $\mu$m, demonstrating a straightforward route to enhance the performance by working on the crystal design
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:34:59 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Zotti",
"Giovanni",
""
],
[
"Horoshko",
"Dmitri B.",
""
],
[
"Kolobov",
"Mikhail I.",
""
],
[
"Michael",
"Yoad",
""
],
[
"Gefen",
"Ziv",
""
],
[
"Chekhova",
"Maria V.",
""
],
[
"Hashimoto",
"Kazuki",
""
]
] |
2509.03350
|
Somiya Chhillar
|
Somiya Chhillar, Mary K. Righi, Rebecca E. Sutter, Evgenios M. Kornaropoulos
|
Exposing Privacy Risks in Anonymizing Clinical Data: Combinatorial Refinement Attacks on k-Anonymity Without Auxiliary Information
| null | null | null | null |
cs.CR
|
http://creativecommons.org/licenses/by/4.0/
|
Despite longstanding criticism from the privacy community, k-anonymity remains a widely used standard for data anonymization, mainly due to its simplicity, regulatory alignment, and preservation of data utility. However, non-experts often defend k-anonymity on the grounds that, in the absence of auxiliary information, no known attacks can compromise its protections. In this work, we refute this claim by introducing Combinatorial Refinement Attacks (CRA), a new class of privacy attacks targeting k-anonymized datasets produced using local recoding. This is the first method that does not rely on external auxiliary information or assumptions about the underlying data distribution. CRA leverages the utility-optimizing behavior of local recoding anonymization of ARX, which is a widely used open-source software for anonymizing data in clinical settings, to formulate a linear program that significantly reduces the space of plausible sensitive values. To validate our findings, we partnered with a network of free community health clinics, an environment where (1) auxiliary information is indeed hard to find due to the population they serve and (2) open-source k-anonymity solutions are attractive due to regulatory obligations and limited resources. Our results on real-world clinical microdata reveal that even in the absence of external information, established anonymization frameworks do not deliver the promised level of privacy, raising critical privacy concerns.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:36:06 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Chhillar",
"Somiya",
""
],
[
"Righi",
"Mary K.",
""
],
[
"Sutter",
"Rebecca E.",
""
],
[
"Kornaropoulos",
"Evgenios M.",
""
]
] |
2509.03351
|
Lloren\c{c} Espinosa-Portal\'es
|
Natalia Flechas Manrique, Alberto Mart\'inez, Elena L\'opez-Mart\'inez, Luc Andrea, Rom\'an Orus, Aitor Manteca, Aitziber L. Cortajarena, Lloren\c{c} Espinosa-Portal\'es
|
epiGPTope: A machine learning-based epitope generator and classifier
|
11 pages, 4 figures. Supplementary Information with 5 pages, 4 figures
| null | null | null |
cs.LG cs.AI q-bio.QM
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Epitopes are short antigenic peptide sequences which are recognized by antibodies or immune cell receptors. These are central to the development of immunotherapies, vaccines, and diagnostics. However, the rational design of synthetic epitope libraries is challenging due to the large combinatorial sequence space, $20^n$ combinations for linear epitopes of n amino acids, making screening and testing unfeasible, even with high throughput experimental techniques. In this study, we present a large language model, epiGPTope, pre-trained on protein data and specifically fine-tuned on linear epitopes, which for the first time can directly generate novel epitope-like sequences, which are found to possess statistical properties analogous to the ones of known epitopes. This generative approach can be used to prepare libraries of epitope candidate sequences. We further train statistical classifiers to predict whether an epitope sequence is of bacterial or viral origin, thus narrowing the candidate library and increasing the likelihood of identifying specific epitopes. We propose that such combination of generative and predictive models can be of assistance in epitope discovery. The approach uses only primary amino acid sequences of linear epitopes, bypassing the need for a geometric framework or hand-crafted features of the sequences. By developing a method to create biologically feasible sequences, we anticipate faster and more cost-effective generation and screening of synthetic epitopes, with relevant applications in the development of new biotechnologies.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:36:06 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Manrique",
"Natalia Flechas",
""
],
[
"Martínez",
"Alberto",
""
],
[
"López-Martínez",
"Elena",
""
],
[
"Andrea",
"Luc",
""
],
[
"Orus",
"Román",
""
],
[
"Manteca",
"Aitor",
""
],
[
"Cortajarena",
"Aitziber L.",
""
],
[
"Espinosa-Portalés",
"Llorenç",
""
]
] |
2509.03352
|
Nero Budur
|
Tom Biesbrouck, Nero Budur, Johannes Nicaise, Willem Veys
|
Birational zeta functions
| null | null | null | null |
math.AG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We define a birational analog of the motivic zeta function of a reduced polynomial in terms of minimal models. It admits an intrinsic meaning in terms of contact loci of arcs, an analog of a result of Denef and Loeser in the motivic case. We show that for local plane curve singularities the poles of the birational zeta function essentially coincide with the poles of the motivic zeta function.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:37:08 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Biesbrouck",
"Tom",
""
],
[
"Budur",
"Nero",
""
],
[
"Nicaise",
"Johannes",
""
],
[
"Veys",
"Willem",
""
]
] |
2509.03353
|
Oguzhan Baser
|
Oguzhan Baser, Kaan Kale, Po-han Li and Sandeep Chinchali
|
Fair Resource Allocation for Fleet Intelligence
|
This paper has been accepted for presentation at the 2025 IEEE Global Communications Conference (GLOBECOM 2025)
| null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Resource allocation is crucial for the performance optimization of cloud-assisted multi-agent intelligence. Traditional methods often overlook agents' diverse computational capabilities and complex operating environments, leading to inefficient and unfair resource distribution. To address this, we open-sourced Fair-Synergy, an algorithmic framework that utilizes the concave relationship between the agents' accuracy and the system resources to ensure fair resource allocation across fleet intelligence. We extend traditional allocation approaches to encompass a multidimensional machine learning utility landscape defined by model parameters, training data volume, and task complexity. We evaluate Fair-Synergy with advanced vision and language models such as BERT, VGG16, MobileNet, and ResNets on datasets including MNIST, CIFAR-10, CIFAR-100, BDD, and GLUE. We demonstrate that Fair-Synergy outperforms standard benchmarks by up to 25% in multi-agent inference and 11% in multi-agent learning settings. Also, we explore how the level of fairness affects the least advantaged, most advantaged, and average agents, providing insights for equitable fleet intelligence.
|
[
{
"version": "v1",
"created": "Tue, 02 Sep 2025 03:20:41 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Baser",
"Oguzhan",
""
],
[
"Kale",
"Kaan",
""
],
[
"Li",
"Po-han",
""
],
[
"Chinchali",
"Sandeep",
""
]
] |
2509.03354
|
Jeremias Resch
|
Jeremias Resch, Ioannis Karapatzakis, Mohamed Elshorbagy, Marcel Schrodin, Philipp Fuchs, Philipp Gra{\ss}hoff, Luis Kussi, Christoph S\"urgers, Cyril Popov, Christoph Becher, Wolfgang Wernsdorfer, David Hunger
|
High-fidelity control of a $^{13}$C nuclear spin coupled to a tin-vacancy center in diamond
|
Main: 5 pages, 3 figures; Supplement: 10 pages, 10 figures, 2 tables
| null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Nuclear spins near group-IV defects in diamond are promising candidates for quantum memories in quantum network applications. Here, we demonstrate high-fidelity control of a single $^{13}$C nuclear spin coupled to a tin-vacancy (SnV) center in diamond. We perform a combination of optical and microwave pumping to achieve initialization into a combined electro-nuclear spin state with a fidelity of $99.74(3)\,\%$. Harnessing a superconducting waveguide for radio-frequency driving, we demonstrate precise nuclear spin control: Ramsey measurements reveal a coherence time of $T_2^* = 1.5(1)\,$milliseconds, and we use dynamical decoupling to extend it to $1.35(3)\,$seconds. We perform randomized benchmarking, yielding a single-qubit gate fidelity of $99.92(1)\,\%$. This demonstrates a coherent spin-photon system with promising properties for quantum network nodes.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:39:14 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Resch",
"Jeremias",
""
],
[
"Karapatzakis",
"Ioannis",
""
],
[
"Elshorbagy",
"Mohamed",
""
],
[
"Schrodin",
"Marcel",
""
],
[
"Fuchs",
"Philipp",
""
],
[
"Graßhoff",
"Philipp",
""
],
[
"Kussi",
"Luis",
""
],
[
"Sürgers",
"Christoph",
""
],
[
"Popov",
"Cyril",
""
],
[
"Becher",
"Christoph",
""
],
[
"Wernsdorfer",
"Wolfgang",
""
],
[
"Hunger",
"David",
""
]
] |
2509.03355
|
BoYe Sun
|
Bo-Ye Sun, Baptiste Bermond, Lucila Peralta Gavensky, Marin Bukov, Zheng-Wei Zhou, Nathan Goldman
|
Controlled Buildup of Half-Quantized Thermal Conductance in an Engineered Chiral Spin Liquid Platform
| null | null | null | null |
cond-mat.quant-gas cond-mat.mes-hall cond-mat.other cond-mat.str-el quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study thermal transport along the edge of a small chiral-spin-liquid device coupled to two Ising-chain reservoirs, a platform suitable for quantum-engineered systems. Adiabatically switching on the tunnel couplings to the reservoirs generates a thermal current that dynamically builds up and reaches a quasi-steady-state regime. In this time window, the two-terminal thermal conductance can approach half-quantized values -- a hallmark of Majorana-mediated transport -- under finely tuned conditions. The results agree with a steady-state Landauer-B\"uttiker description for sufficiently large reservoirs, where energy-resolved transmission rates help identify the optimal parameters to achieve the half-quantized conductance. This work provides a controllable platform to investigate topological thermal transport in engineered spin systems, such as realized in cold-atom and Rydberg-atom settings.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:42:13 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Sun",
"Bo-Ye",
""
],
[
"Bermond",
"Baptiste",
""
],
[
"Gavensky",
"Lucila Peralta",
""
],
[
"Bukov",
"Marin",
""
],
[
"Zhou",
"Zheng-Wei",
""
],
[
"Goldman",
"Nathan",
""
]
] |
2509.03356
|
Mykola Isaiev
|
Viktor Mandrolko and Mykola Isaiev
|
Tailored Thermal Transport in Phase Change Materials-Based Nanocomposites through Interfacial Structuring
| null | null | null | null |
cond-mat.mes-hall physics.app-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Interfacial thermal transport is a critical bottleneck in nanoscale systems, where heat dissipation and energy efficiency are strongly modulated by molecular ordering at solid-liquid boundaries. Here, using atomistic simulations of hexadecane confined by structured silica substrates, we reveal how interfacial geometry, specifically curvature, governs the density distribution and thermal transport across the interface. At flat and mildly curved surfaces, the liquid exhibits surface-templated layering, promoting efficient heat transfer, which is enhanced with increasing contact surface area. As curvature increases, this ordering breaks down, giving rise to interference-like density patterns, reduced molecular packing, and localized depletion zones. This structural reorganization leads to a systematic increase in interfacial thermal resistance (ITR), even when the contact area is kept constant. By decomposing the interface into convex ("hill" of solid) and concave ("valley" of solid) regions, we find that valleys consistently offer lower thermal resistance. In contrast, hills act as bottlenecks to heat flow. Remarkably, we show that the work of adhesion and entropy-related energy losses scale non-trivially with curvature: while adhesion increases with contact area, the entropic penalty dominates the total energy change, reflecting curvature-induced frustration of molecular alignment. These findings unveil a direct link between surface geometry, thermodynamic dissipation, and heat transport, offering new design principles for thermally tunable nanostructured materials, thermal interface coatings, and phase-change systems.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:42:16 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Mandrolko",
"Viktor",
""
],
[
"Isaiev",
"Mykola",
""
]
] |
2509.03357
|
Alexia Ambrogio
|
Alexia Ambrogio, Leonardo Cianfanelli, Giacomo Como
|
On Optimality of Private Information in Bayesian Routing Games
| null | null | null | null |
math.OC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study an information design problem in transportation networks, in the presence of a random state that affects the travel times on the links. An omniscient system planner -- aiming at reducing congestion -- observes the network state realization and sends private messages to the users -- who share a common prior on the network state but do not observe it directly -- in order to nudge them towards a socially desirable behavior. The desired effect of these private signals is to correlate the users' selfish decisions with the network state and align the resulting Bayesian Wardrop equilibrium with the system optimum flow. Our main contribution is to provide sufficient and necessary conditions under which optimality may be achieved by a fair private signal policy in transportation networks with injective link-path incidence matrix and affine travel time functions.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:43:14 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Ambrogio",
"Alexia",
""
],
[
"Cianfanelli",
"Leonardo",
""
],
[
"Como",
"Giacomo",
""
]
] |
2509.03358
|
Bhanu Sharma Mr.
|
Bhanu Sharma and Eugene Pinsky
|
Some patterns of sleep quality and Daylight Saving Time across countries: a predictive and exploratory analysis
|
16 Pages
|
International Journal of Data Mining & Knowledge Management Process (IJDKP) 2025
|
10.5121/ijdkp.2025.15401
|
Vol 15, No. 4
|
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
In this study we analyzed average sleep durations across 61 countries to examine the impact of Daylight Saving Time (DST) practices. Key metrics influencing sleep were identified, and statistical correlation analysis was applied to explore relationships among these factors. Countries were grouped based on DST observance, and visualizations compared sleep patterns between DST and non-DST regions. Results show that, on average, countries observing DST tend to report longer sleep durations than those that do not. A more detailed pattern emerged when accounting for latitude: at lower latitudes, DST-observing countries reported shorter sleep durations compared to non-DST countries, while at higher latitudes, DST-observing countries reported longer average sleep durations. These findings suggest that the influence of DST on sleep may be moderated by geographical location.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:44:18 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Sharma",
"Bhanu",
""
],
[
"Pinsky",
"Eugene",
""
]
] |
2509.03359
|
Aman Anand
|
A. Anand, R. Egger, A. De Martino
|
Magnetic Bloch bands and Weiss oscillations in Dirac mass superlattices
|
16 pages, 10 figures
| null | null | null |
cond-mat.mes-hall
|
http://creativecommons.org/licenses/by/4.0/
|
We study two-dimensional Dirac fermions in a one-dimensional mass superlattice under a perpendicular magnetic field. Using exact solutions for isolated and finite arrays of domain walls, we demonstrate the persistence of Jackiw-Rebbi modes with a field-dependent renormalized velocity. For the periodic case, we adopt a gauge-invariant projection method onto magnetic Bloch states, valid for arbitrary fields and mass profiles, which yields dispersive Landau levels, and confirm its accuracy by comparison with finite arrays spectra. From the miniband spectra we predict modified quantum Hall plateaus and Weiss-like magnetoconductivity oscillations, characterized by a strongly reduced amplitude and a $\pi/2$ phase shift compared to electrostatic superlattices.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:45:37 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Anand",
"A.",
""
],
[
"Egger",
"R.",
""
],
[
"De Martino",
"A.",
""
]
] |
2509.03360
|
Christopher Berry
|
S Al-Shammari, C P L Berry, C E A Chapman-Bird, F T Chowdhury, K Cunningham, M Emma, R Gray, C Hoy, I S Heng, M Korobko, E Maggio, A-K Malz, H Middleton, M Prathaban, I M Romero-Shaw, G Shaifullah, S Singh, J Steinlechner, K Toland, D Williams, and M J Williams
|
The Early Career Workshop of GR-Amaldi 2025
|
9 pages 1 table, 1 figures. Submitted to the proceedings of the GR24-Amaldi16 meeting
| null | null | null |
physics.soc-ph gr-qc
|
http://creativecommons.org/licenses/by/4.0/
|
Gravitational physics and astronomy have developed rapidly over the last decade, driven by new observations and theoretical breakthroughs. As new as the science and technology of this field are, its greatest asset may be the body of early-career researchers actively engaged in driving it forward. With the aim of bringing together this community of enthusiastic scientists from a broad array of disciplines, the organisers of the GR-Amaldi meeting joined with the Gravitational-Wave Early Career Scientists organisation to create a three-day event: the Early Career Workshop. The Workshop aimed to provide a broad overview of the field's diverse scientific possibilities and introduce key theoretical foundations underpinning its science. To complement developing technical skills, the Workshop also sought to provide participants with transferable skills to aid them in their future careers. The Workshop emphasized networking and community building, offering participants opportunities to engage with peers and mentors. It encouraged interdisciplinary exchanges and cross-institutional collaboration, fostering connections across different research efforts. Collectively, these initiatives aimed to equip participants with a comprehensive understanding of the field's research and to build a more cohesive, collaborative community of early-career researchers. We summarise key points and conclusion from the carious activities carried out as part of the Workshop.
|
[
{
"version": "v1",
"created": "Sat, 30 Aug 2025 17:29:12 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Al-Shammari",
"S",
""
],
[
"Berry",
"C P L",
""
],
[
"Chapman-Bird",
"C E A",
""
],
[
"Chowdhury",
"F T",
""
],
[
"Cunningham",
"K",
""
],
[
"Emma",
"M",
""
],
[
"Gray",
"R",
""
],
[
"Hoy",
"C",
""
],
[
"Heng",
"I S",
""
],
[
"Korobko",
"M",
""
],
[
"Maggio",
"E",
""
],
[
"Malz",
"A-K",
""
],
[
"Middleton",
"H",
""
],
[
"Prathaban",
"M",
""
],
[
"Romero-Shaw",
"I M",
""
],
[
"Shaifullah",
"G",
""
],
[
"Singh",
"S",
""
],
[
"Steinlechner",
"J",
""
],
[
"Toland",
"K",
""
],
[
"Williams",
"D",
""
],
[
"Williams",
"M J",
""
]
] |
2509.03361
|
Yiyang Guo
|
Yiyang Guo, Luyao Zhang, Fabo Feng, Zhao-Yu Li, Anton Pomazan, Xiaohu Yang
|
Search for Past Stellar Encounters and the Origin of 3I/ATLAS
|
11 pages, 5 figures, submitted to ApJL
| null | null | null |
astro-ph.SR astro-ph.EP astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
3I/ATLAS, the third discovered interstellar object, has a heliocentric speed of 58 km/s and exhibits cometary activity. To constrain the origin of 3I/ATLAS and its past dynamical evolution, we propagate the orbits of 3I/ATLAS and nearby stars to search for stellar encounters. Integrating orbits in the Galactic potential and propagating the astrometric and radial-velocity uncertainties of 30 million Gaia stars, we identify 25 encounters with median encounter distances less than 1 pc. However, because the encounter speeds between 3I/ATLAS and each encounter exceed 20 km/s, none is a plausible host under common ejection mechanisms. We infer stellar masses for most stars and quantify the gravitational perturbations exerted by each individual star or each binary system on 3I/ATLAS. The strongest gravitational scattering perturber is a wide M-dwarf binary. Among all past encounters, the binary's barycenter and 3I/ATLAS reach the small encounter distance of 0.242 pc and the encounter speed of 28.39 km/s,1.64 Myr ago. We further demonstrate that the cumulative influence of the stellar encounters on both the speed and direction of 3I/ATLAS is weak. Based on the present kinematics of 3I/ATLAS to assess its origin, we find that a thin-disk origin is strongly favored, because the thin disk both exhibits a velocity distribution closely matching that of 3I/ATLAS and provides the dominant local number density of stars.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:46:05 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Guo",
"Yiyang",
""
],
[
"Zhang",
"Luyao",
""
],
[
"Feng",
"Fabo",
""
],
[
"Li",
"Zhao-Yu",
""
],
[
"Pomazan",
"Anton",
""
],
[
"Yang",
"Xiaohu",
""
]
] |
2509.03362
|
Pedro Alvarez
|
P. D. Alvarez, J. Ortiz and J. Zanelli
|
Gravitating spinor torsional BTZ solution
|
20 pages
| null | null | null |
hep-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We construct exact analytic solutions for a supersymmetric Chern-Simons theory with matter fields in the adjoint representation (unconventional supersymmetry). The configurations correspond to gravitating spinor fields on BTZ geometries with torsion. Although the spinor profiles exhibit divergent behavior near the black hole horizon, they yield finite contributions to conserved charges. We demonstrate that the spinors solve the full nonlinear field equations and satisfy integrability conditions derived from the Bianchi identities. The resulting energy momentum tensor is traceless, consistent with the Weyl invariance of the theory. We also analyze spinor solutions based on $adS_3$ Killing spinors and identify the conditions under which such backgrounds can support nontrivial spinor configurations. Our findings open the way to new fermionic sectors in 3D gravity with torsion, distinct from previously known perturbative deformations.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:46:42 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Alvarez",
"P. D.",
""
],
[
"Ortiz",
"J.",
""
],
[
"Zanelli",
"J.",
""
]
] |
2509.03363
|
Denislav Nikolov
|
Denislav P. Nikolov, Zhiren Zhu, Jonathan B. Estrada
|
Variation-matching sensitivity-based virtual fields for hyperelastic material model calibration
| null | null | null | null |
cond-mat.soft
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Accurate identification of nonlinear material parameters from three-dimensional full-field deformation data remains a challenge in experimental mechanics. The virtual fields method (VFM) provides a powerful, computationally efficient approach for material model calibration, however, its success depends critically on the choice of virtual fields and the informativeness of available kinematic data. In this work, we advance the state-of-the-art discrete formulation of the sensitivity-based virtual fields (SBVF) method by systematically developing and comparing alternative variational and analytical SBVFs within a strain-invariant-based modeling framework.
A central contribution of this work is the implementation and assessment of variation-based SBVFs (vSBVFs), formulated using directional G\^ateaux derivatives, as well as virtual fields derived from analytical differentiation (aSBVFs) which provide explicit, model-tailored virtual displacement fields for parameter identification. Using simulated noisy volumetric datasets, we demonstrate that vSBVFs and aSBVFs enable procedural, automated construction of optimal virtual fields for each material parameter, substantially enhancing the robustness and efficiency of calibration without the need for manual field selection or high temporal resolution in the data acquisition. We quantify data richness -- the effective diversity of sampled kinematic states -- showing that increased data richness via sample geometry and loading protocols leads to improved parameter identifiability. These findings establish a pathway for automated, noise-robust material model calibration suitable for future deployment with experimental full-field imaging of soft, complex materials, and provide a foundation for optimizing shape topology and extending to viscoelastic and anisotropic behaviors.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:46:57 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Nikolov",
"Denislav P.",
""
],
[
"Zhu",
"Zhiren",
""
],
[
"Estrada",
"Jonathan B.",
""
]
] |
2509.03364
|
Hai Ngo Ngoc
|
Hai N Ngo, Dieu D. Nguyen, Truong N. Nguyen, Trung H. Dang, Tien H.T. Ho
|
Extending the simulations of intermediate-mass black hole mass measurements to Virgo Cluster using ELT/HARMONI high resolution integral-field stellar kinematics
| null | null | null | null |
astro-ph.GA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The firm co-existence of intermediate-mass black holes (IMBHs, $M_{\rm BH} \approx 10^3-10^6 M_odot$) in nuclear star clusters (NSCs) remains uncertain because the limited number of verified instances within the local Universe, limited within 3.5 Mpc. They are crucial for our understanding about the formation and evolution of supermassive black holes (SMBHs). The upcoming Extremely Large Telescope (ELT) promises to revolutionize the detection of these mysterious objects. In this study, we simulated the kinematics of an IMBH within the nuclear star cluster of VCC 1861, one of the faintest galaxies in the Virgo Cluster. Using Jeans Anisotropic Modeling (JAM) and stellar density profiles derived from Hubble Space Telescope (HST) data and the HSIM program, we created mock High Angular Resolution Monolithic Optical and Near-infrared Integral field spectrograph (HARMONI) observations for the ELT. We then extract stellar kinematics from these mock data and recover the BH mass using the JAM model with Markov Chain Monte Carlo simulation method. Our results demonstrate the ELT's capability to detect IMBHs with masses comprising 5% of the NSC's mass at the distance of the Virgo Cluster.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:48:04 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Ngo",
"Hai N",
""
],
[
"Nguyen",
"Dieu D.",
""
],
[
"Nguyen",
"Truong N.",
""
],
[
"Dang",
"Trung H.",
""
],
[
"Ho",
"Tien H. T.",
""
]
] |
2509.03365
|
Paul-Gauthier No\'e
|
Paul-Gauthier No\'e, Andreas Nautsch, Driss Matrouf, Pierre-Michel Bousquet, Jean-Fran\c{c}ois Bonastre
|
The distribution of calibrated likelihood functions on the probability-likelihood Aitchison simplex
|
Preprint. Under review
| null | null | null |
cs.LG stat.ML
|
http://creativecommons.org/licenses/by/4.0/
|
While calibration of probabilistic predictions has been widely studied, this paper rather addresses calibration of likelihood functions. This has been discussed, especially in biometrics, in cases with only two exhaustive and mutually exclusive hypotheses (classes) where likelihood functions can be written as log-likelihood-ratios (LLRs). After defining calibration for LLRs and its connection with the concept of weight-of-evidence, we present the idempotence property and its associated constraint on the distribution of the LLRs. Although these results have been known for decades, they have been limited to the binary case. Here, we extend them to cases with more than two hypotheses by using the Aitchison geometry of the simplex, which allows us to recover, in a vector form, the additive form of the Bayes' rule; extending therefore the LLR and the weight-of-evidence to any number of hypotheses. Especially, we extend the definition of calibration, the idempotence, and the constraint on the distribution of likelihood functions to this multiple hypotheses and multiclass counterpart of the LLR: the isometric-log-ratio transformed likelihood function. This work is mainly conceptual, but we still provide one application to machine learning by presenting a non-linear discriminant analysis where the discriminant components form a calibrated likelihood function over the classes, improving therefore the interpretability and the reliability of the method.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:48:11 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Noé",
"Paul-Gauthier",
""
],
[
"Nautsch",
"Andreas",
""
],
[
"Matrouf",
"Driss",
""
],
[
"Bousquet",
"Pierre-Michel",
""
],
[
"Bonastre",
"Jean-François",
""
]
] |
2509.03366
|
Anika Kumar
|
Anika Kumar, David J. Setton, Rachel Bezanson, Alan Pearl, Erin Stumbaugh, Justin S. Spilker, Vincenzo R. D'Onofrio, Jenny E. Greene, Katherine A. Suess, and Margaret E. Verrico
|
Meet the Neighbors: Gas Rich "Buddy Galaxies" are Common Around Recently Quenched Massive Galaxies in the SQuIGG$\vec{L}$E Survey
|
Submitted to AAS;4 pages, 1 figure. This work is complemented by a companion paper, Setton et al., which presents the molecular gas properties of the host post-starburst galaxies
| null | null | null |
astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
In this work, we characterize the environments of massive ($\log(M_\odot/M_\star)\sim11.2$) $z\sim0.7$ post-starburst galaxies (PSBs) by studying serendipitously-detected CO(2-1) emitters found in targeted observations of the SQuIGG$\vec{L}$E sample. We report $31\pm6\%$ of the galaxies from this survey host nearby gas-rich ``buddies'' with stellar masses $\geq 10^{10},M_\odot$ and molecular gas comparable to their central PSBs ($M_{H_{2}} \sim 10^{10} M_\odot$), but $\sim0.8$ dex lower stellar mass ($\sim 10^{10.4} M_\odot$). Based on their location in position-velocity space, each buddy is consistent with being bound to the haloes of their SQuIGG$\vec{L}$E host galaxies. We compare to the UniverseMachine model and find that SQuIGG$\vec{L}$E galaxies host a typical number of neighbors for their stellar mass, suggesting that PSBs live in environments typical of co-eval similarly-massive galaxies.
|
[
{
"version": "v1",
"created": "Fri, 29 Aug 2025 18:00:00 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Kumar",
"Anika",
""
],
[
"Setton",
"David J.",
""
],
[
"Bezanson",
"Rachel",
""
],
[
"Pearl",
"Alan",
""
],
[
"Stumbaugh",
"Erin",
""
],
[
"Spilker",
"Justin S.",
""
],
[
"D'Onofrio",
"Vincenzo R.",
""
],
[
"Greene",
"Jenny E.",
""
],
[
"Suess",
"Katherine A.",
""
],
[
"Verrico",
"Margaret E.",
""
]
] |
2509.03367
|
Narges Dadkhah
|
Narges Dadkhah, Somayeh Mohammadi, Gerhard Wunder
|
Tuning Block Size for Workload Optimization in Consortium Blockchain Networks
| null | null |
10.1109/ICBC64466.2025.11114510
| null |
cs.CR
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Determining the optimal block size is crucial for achieving high throughput in blockchain systems. Many studies have focused on tuning various components, such as databases, network bandwidth, and consensus mechanisms. However, the impact of block size on system performance remains a topic of debate, often resulting in divergent views and even leading to new forks in blockchain networks. This research proposes a mathematical model to maximize performance by determining the ideal block size for Hyperledger Fabric, a prominent consortium blockchain. By leveraging machine learning and solving the model with a genetic algorithm, the proposed approach assesses how factors such as block size, transaction size, and network capacity influence the block processing time. The integration of an optimization solver enables precise adjustments to block size configuration before deployment, ensuring improved performance from the outset. This systematic approach aims to balance block processing efficiency, network latency, and system throughput, offering a robust solution to improve blockchain performance across diverse business contexts.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 18:03:00 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Dadkhah",
"Narges",
""
],
[
"Mohammadi",
"Somayeh",
""
],
[
"Wunder",
"Gerhard",
""
]
] |
2509.03368
|
Edoardo Alviani
|
Edoardo Alviani and Adam Falkowski
|
Gauge anomalies on shell and collinear factorization
|
17 pages, 2 figures
| null | null | null |
hep-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We revisit gauge anomalies from the purely on-shell perspective. We argue that violation of anomaly cancellation conditions manifests as a breakdown of collinear factorization. We explicitly construct one-loop 5-point amplitudes with graviton exchange displaying singular behavior in collinear limits that cannot be reconciled with factorization theorems. In this approach, gravity serves as a universal probe of both abelian and non-abelian anomalies.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:49:42 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Alviani",
"Edoardo",
""
],
[
"Falkowski",
"Adam",
""
]
] |
2509.03369
|
Alexander Bednyakov
|
A.V. Bednyakov, A.S. Fedoruk, D.I. Kazakov
|
On the renormalization-group analysis of the SM: loops, uncertainties, and vacuum stability
|
37 pages, 24 figures
| null | null | null |
hep-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Renormalization-group equations (RGE) is one of the key tools in studying high-energy behavior of the Standard Model (SM). We begin by reviewing one-loop RGE for the dimensionless couplings of the SM and proceed to the state-of-the-art results. Our study focuses on the RGE solutions at different loop orders. We compare not only the standard (``diagonal'') loop counting, when one considers gauge, Yukawa, and scalar self-coupling beta functions at the same order, but also ``non-diagonal'' ones, inspired by the so-called Weyl consistency conditions. We discuss the initial conditions for RGE (``matching'') for different loop configurations, and study the uncertainties of running coupling both related to the limited precision of the experimental input (``parametric'') and to the missing high-order corrections (``theoretical''). As an application of our analysis we also estimate the electroweak vacuum decay probability and study how the uncertainties in the running parameters affect the latter. We argue that the ``non-diagonal'' beta functions, if coupled with more consistent ``non-diagonal'' matching lead to larger theoretical uncertainty than the ``diagonal'' ones.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:50:01 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Bednyakov",
"A. V.",
""
],
[
"Fedoruk",
"A. S.",
""
],
[
"Kazakov",
"D. I.",
""
]
] |
2509.03370
|
Akash Malhotra
|
Akash Malhotra, Nac\'era Seghouani
|
Neural Field Turing Machine: A Differentiable Spatial Computer
|
11 Pages, 6 Figures
| null | null | null |
cs.NE cs.AI
|
http://creativecommons.org/publicdomain/zero/1.0/
|
We introduce the Neural Field Turing Machine (NFTM), a differentiable architecture that unifies symbolic computation, physical simulation, and perceptual inference within continuous spatial fields. NFTM combines a neural controller, continuous memory field, and movable read/write heads that perform local updates. At each timestep, the controller reads local patches, computes updates via learned rules, and writes them back while updating head positions. This design achieves linear O(N) scaling through fixed-radius neighborhoods while maintaining Turing completeness under bounded error. We demonstrate three example instantiations of NFTM: cellular automata simulation (Rule 110), physics-informed PDE solvers (2D heat equation), and iterative image refinement (CIFAR-10 inpainting). These instantiations learn local update rules that compose into global dynamics, exhibit stable long-horizon rollouts, and generalize beyond training horizons. NFTM provides a unified computational substrate bridging discrete algorithms and continuous field dynamics within a single differentiable framework.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 22:29:15 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Malhotra",
"Akash",
""
],
[
"Seghouani",
"Nacéra",
""
]
] |
2509.03371
|
El Mehdi Cherradi
|
El Mehdi Cherradi
|
Internal languages of locally cartesian closed $(\infty,1)$-categories
|
65 pages, comments welcome!
| null | null | null |
math.CT math.AT math.LO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We establish a DK-equivalence between the relative category of $\pi$-tribes and the relative category of locally cartesian closed quasicategories. From this follows one of the internal languages conjecture: Martin-L\"of type theory with dependent sums, intensional identity types, and dependent products satisfying functional extensionality is the internal language of locally cartesian closed $(\infty,1)$-categories.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:51:19 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Cherradi",
"El Mehdi",
""
]
] |
2509.03372
|
Tien-Hong Lo
|
Tien-Hong Lo, Szu-Yu Chen, Yao-Ting Sung, Berlin Chen
|
An Effective Strategy for Modeling Score Ordinality and Non-uniform Intervals in Automated Speaking Assessment
|
Accepted at ASRU 2025
| null | null | null |
eess.AS cs.LG cs.SD
|
http://creativecommons.org/licenses/by/4.0/
|
A recent line of research on automated speaking assessment (ASA) has benefited from self-supervised learning (SSL) representations, which capture rich acoustic and linguistic patterns in non-native speech without underlying assumptions of feature curation. However, speech-based SSL models capture acoustic-related traits but overlook linguistic content, while text-based SSL models rely on ASR output and fail to encode prosodic nuances. Moreover, most prior arts treat proficiency levels as nominal classes, ignoring their ordinal structure and non-uniform intervals between proficiency labels. To address these limitations, we propose an effective ASA approach combining SSL with handcrafted indicator features via a novel modeling paradigm. We further introduce a multi-margin ordinal loss that jointly models both the score ordinality and non-uniform intervals of proficiency labels. Extensive experiments on the TEEMI corpus show that our method consistently outperforms strong baselines and generalizes well to unseen prompts.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 09:18:51 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Lo",
"Tien-Hong",
""
],
[
"Chen",
"Szu-Yu",
""
],
[
"Sung",
"Yao-Ting",
""
],
[
"Chen",
"Berlin",
""
]
] |
2509.03373
|
Elizabeth Coda
|
Elizabeth Coda and Ery Arias-Castro and Gal Mishne
|
Cluster and then Embed: A Modular Approach for Visualization
| null | null | null | null |
cs.LG stat.ME stat.ML
|
http://creativecommons.org/licenses/by/4.0/
|
Dimensionality reduction methods such as t-SNE and UMAP are popular methods for visualizing data with a potential (latent) clustered structure. They are known to group data points at the same time as they embed them, resulting in visualizations with well-separated clusters that preserve local information well. However, t-SNE and UMAP also tend to distort the global geometry of the underlying data. We propose a more transparent, modular approach consisting of first clustering the data, then embedding each cluster, and finally aligning the clusters to obtain a global embedding. We demonstrate this approach on several synthetic and real-world datasets and show that it is competitive with existing methods, while being much more transparent.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 00:27:30 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Coda",
"Elizabeth",
""
],
[
"Arias-Castro",
"Ery",
""
],
[
"Mishne",
"Gal",
""
]
] |
2509.03374
|
Laura Schaposnik
|
Alyssa Yu and Laura P. Schaposnik
|
Dynamics of Infection Spread and Hotspot Growth in Bi-Pathogen Networks
|
20 pages, 17 figures, 2 tables comments welcomed!
| null | null | null |
physics.soc-ph math.DS nlin.AO q-bio.PE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Understanding the spatio-temporal evolution of epidemics with multiple pathogens requires not only new theoretical models but also careful analysis of their practical consequences. Building on the Multiplex Bi-Virus Reaction-Diffusion framework (MBRD) introduced in our companion paper, we investigate how the super-infection model (MBRD-SI) and the co-infection model (MBRD-CI) behave under different epidemiological and network conditions. Through numerical experiments, we study the effects of pathogen virulence, diffusion rates, and cross-diffusion on epidemic hotspot formation and long-term prevalence. Our results highlight the role of multiplex structure in amplifying or suppressing co-circulating infections, and provide quantitative insight into conditions that drive persistent epidemic patterns. Beyond epidemiology, these findings have broader implications for multiplex contagion processes such as information diffusion and malware propagation.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:52:46 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Yu",
"Alyssa",
""
],
[
"Schaposnik",
"Laura P.",
""
]
] |
2509.03375
|
Martin Jirlow
|
Martin Jirlow, Kunal Helambe, Axel M. Eriksson, Simone Gasparinetti, Tahereh Abad
|
Effective Hamiltonian for an off-resonantly driven qubit-cavity system
|
6+6 pages, 4+2 figures
| null | null | null |
quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Accurate modeling of driven light-matter interactions is essential for quantum technologies, where natural and synthetic atoms are used to store and process quantum information, mediate interactions between bosonic modes, and enable nonlinear operations. In systems subject to multi-tone drives, however, the theoretical description becomes challenging and existing models cannot quantitatively reproduce the experimental data. Here, we derive an effective Hamiltonian that retains slowly rotating terms, providing a general framework for accurately describing driven dynamics across platforms. As a concrete application, we validate the theory in circuit QED, where it quantitatively reproduces experimentally measured ac Stark shifts and captures key interactions such as two-mode squeezing and beam-splitting. Our results establish a broadly applicable tool to engineer driven interactions in quantum information processing platforms.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:53:07 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Jirlow",
"Martin",
""
],
[
"Helambe",
"Kunal",
""
],
[
"Eriksson",
"Axel M.",
""
],
[
"Gasparinetti",
"Simone",
""
],
[
"Abad",
"Tahereh",
""
]
] |
2509.03376
|
Xianchao Xiu
|
Hui Chen, Liangyu Liu, Xianchao Xiu, Wanquan Liu
|
Transformer-Guided Content-Adaptive Graph Learning for Hyperspectral Unmixing
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Hyperspectral unmixing (HU) targets to decompose each mixed pixel in remote sensing images into a set of endmembers and their corresponding abundances. Despite significant progress in this field using deep learning, most methods fail to simultaneously characterize global dependencies and local consistency, making it difficult to preserve both long-range interactions and boundary details. This letter proposes a novel transformer-guided content-adaptive graph unmixing framework (T-CAGU), which overcomes these challenges by employing a transformer to capture global dependencies and introducing a content-adaptive graph neural network to enhance local relationships. Unlike previous work, T-CAGU integrates multiple propagation orders to dynamically learn the graph structure, ensuring robustness against noise. Furthermore, T-CAGU leverages a graph residual mechanism to preserve global information and stabilize training. Experimental results demonstrate its superiority over the state-of-the-art methods. Our code is available at https://github.com/xianchaoxiu/T-CAGU.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:53:33 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Chen",
"Hui",
""
],
[
"Liu",
"Liangyu",
""
],
[
"Xiu",
"Xianchao",
""
],
[
"Liu",
"Wanquan",
""
]
] |
2509.03377
|
Rui Xie
|
Rui Xie, Asad Ul Haq, Linsen Ma, Yunhua Fang, Zirak Burzin Engineer, Liu Liu, Tong Zhang
|
Amplifying Effective CXL Memory Bandwidth for LLM Inference via Transparent Near-Data Processing
| null | null | null | null |
cs.AR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Large language model (LLM) inference is bottlenecked by the limited bandwidth of CXL-based memory used for capacity expansion. We introduce CXL-NDP, a transparent near-data processing architecture that amplifies effective CXL bandwidth without requiring changes to the CXL.mem interface or AI models. CXL-NDP integrates a precision-scalable bit-plane layout for dynamic quantization with transparent lossless compression of weights and KV caches directly within the CXL device. In end-to-end serving, CXL-NDP improves throughput by 43%, extends the maximum context length by 87%, and reduces the KV cache footprint by 46.9% without accuracy loss. Hardware synthesis confirms its practicality with a modest silicon footprint, lowering the barrier for adopting efficient, scalable CXL-based memory in generative AI infrastructure.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:53:45 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Xie",
"Rui",
""
],
[
"Haq",
"Asad Ul",
""
],
[
"Ma",
"Linsen",
""
],
[
"Fang",
"Yunhua",
""
],
[
"Engineer",
"Zirak Burzin",
""
],
[
"Liu",
"Liu",
""
],
[
"Zhang",
"Tong",
""
]
] |
2509.03378
|
Wu Lin
|
Wu Lin, Scott C. Lowe, Felix Dangel, Runa Eschenhagen, Zikun Xu, Roger B. Grosse
|
Understanding and Improving the Shampoo Optimizer via Kullback-Leibler Minimization
|
technical report, working in progress
| null | null | null |
stat.ML cs.LG
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
As an adaptive method, Shampoo employs a structured second-moment estimation, and its effectiveness has attracted growing attention. Prior work has primarily analyzed its estimation scheme through the Frobenius norm. Motivated by the natural connection between the second moment and a covariance matrix, we propose studying Shampoo's estimation as covariance estimation through the lens of Kullback-Leibler (KL) minimization. This alternative perspective reveals a previously hidden limitation, motivating improvements to Shampoo's design. Building on this insight, we develop a practical estimation scheme, termed KL-Shampoo, that eliminates Shampoo's reliance on Adam for stabilization, thereby removing the additional memory overhead introduced by Adam. Preliminary results show that KL-Shampoo improves Shampoo's performance, enabling it to stabilize without Adam and even outperform its Adam-stabilized variant, SOAP, in neural network pretraining.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:55:15 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Lin",
"Wu",
""
],
[
"Lowe",
"Scott C.",
""
],
[
"Dangel",
"Felix",
""
],
[
"Eschenhagen",
"Runa",
""
],
[
"Xu",
"Zikun",
""
],
[
"Grosse",
"Roger B.",
""
]
] |
2509.03379
|
Guoxin Wang
|
Guoxin Wang, Qingyuan Wang, Binhua Huang, Shaowu Chen, Deepu John
|
TinyDrop: Tiny Model Guided Token Dropping for Vision Transformers
| null | null | null | null |
cs.CV cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Vision Transformers (ViTs) achieve strong performance in image classification but incur high computational costs from processing all image tokens. To reduce inference costs in large ViTs without compromising accuracy, we propose TinyDrop, a training-free token dropping framework guided by a lightweight vision model. The guidance model estimates the importance of tokens while performing inference, thereby selectively discarding low-importance tokens if large vit models need to perform attention calculations. The framework operates plug-and-play, requires no architectural modifications, and is compatible with diverse ViT architectures. Evaluations on standard image classification benchmarks demonstrate that our framework reduces FLOPs by up to 80% for ViTs with minimal accuracy degradation, highlighting its generalization capability and practical utility for efficient ViT-based classification.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:55:49 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Wang",
"Guoxin",
""
],
[
"Wang",
"Qingyuan",
""
],
[
"Huang",
"Binhua",
""
],
[
"Chen",
"Shaowu",
""
],
[
"John",
"Deepu",
""
]
] |
2509.03380
|
Dr Peter J. Bentley
|
Peter J. Bentley, Soo Ling Lim, and Fuyuki Ishikawa
|
Situating AI Agents in their World: Aspective Agentic AI for Dynamic Partially Observable Information Systems
|
9 pages
|
7th International Workshop on Agent-Based Modelling of Human Behaviour (ABMHuB'25), ALife 2025
| null | null |
cs.AI cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Agentic LLM AI agents are often little more than autonomous chatbots: actors following scripts, often controlled by an unreliable director. This work introduces a bottom-up framework that situates AI agents in their environment, with all behaviors triggered by changes in their environments. It introduces the notion of aspects, similar to the idea of umwelt, where sets of agents perceive their environment differently to each other, enabling clearer control of information. We provide an illustrative implementation and show that compared to a typical architecture, which leaks up to 83% of the time, aspective agentic AI enables zero information leakage. We anticipate that this concept of specialist agents working efficiently in their own information niches can provide improvements to both security and efficiency.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:57:04 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Bentley",
"Peter J.",
""
],
[
"Lim",
"Soo Ling",
""
],
[
"Ishikawa",
"Fuyuki",
""
]
] |
2509.03381
|
Sanghoon Lee Mr.
|
Sanghoon Lee, Junha Kang, Kyung-Joon Park
|
Dependency Chain Analysis of ROS 2 DDS QoS Policies: From Lifecycle Tutorial to Static Verification
|
14 pages, 4 figures
| null | null | null |
cs.NI cs.RO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Robot Operating System 2 (ROS 2) relies on the Data Distribution Service (DDS), which offers more than 20 Quality of Service (QoS) policies governing availability, reliability, and resource usage. Yet ROS 2 users lack clear guidance on safe policy combinations and validation processes prior to deployment, which often leads to trial-and-error tuning and unexpected runtime failures. To address these challenges, we analyze DDS Publisher-Subscriber communication over a life cycle divided into Discovery, Data Exchange, and Disassociation, and provide a user oriented tutorial explaining how 16 QoS policies operate in each phase. Building on this analysis, we derive a QoS dependency chain that formalizes inter-policy relationships and classifies 41 dependency violation rules, capturing constraints that commonly cause communication failures in practice. Finally, we introduce QoS Guard, a ROS 2 package that statically validates DDS XML profiles offline, flags conflicts, and enables safe, predeployment tuning without establishing a live ROS 2 session. Together, these contributions give ROS 2 users both conceptual insight and a concrete tool that enables early detection of misconfigurations, improving the reliability and resource efficiency of ROS 2 based robotic systems.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:57:21 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Lee",
"Sanghoon",
""
],
[
"Kang",
"Junha",
""
],
[
"Park",
"Kyung-Joon",
""
]
] |
2509.03382
|
Arijit Manna
|
Arijit Manna, Sabyasachi Pal
|
First detection of ethylene oxide and acetaldehyde in hot core G358.93$-$0.03 MM1: Tracing prebiotic oxygen chemistry
|
Published in New Astronomy
|
New Astron, 122, 102465 (2026)
|
10.1016/j.newast.2025.102465
| null |
astro-ph.GA physics.chem-ph physics.space-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Ethylene oxide (c-C$_{2}$H$_{4}$O) and its isomer, acetaldehyde (CH$_{3}$CHO), are important complex organic molecules owing to their potential role in the formation of amino acids (R-CH(NH$_{2}$)-COOH) in ISM. The detection of c-C$_{2}$H$_{4}$O in hot molecular cores suggests that the possible existence of larger ring-shaped molecules containing more than three carbon atoms, such as furan (c-C$_{4}$H$_{4}$O), which shares structural similarities with ribose (C$_{5}$H$_{10}$O5), the sugar component of DNA. In this study, we report the first detection of the rotational emission lines of c-C$_{2}$H$_{4}$O and CH$_{3}$CHO towards the hot molecular core G358.93$-$0.03 MM1, based on observations from the Atacama Large Millimeter/Submillimeter Array (ALMA) in band 7. The fractional abundances of c-C$_{2}$H$_{4}$O and CH$_{3}$CHO relative to H$_{2}$ are $(2.1\pm0.2)\times10^{-9}$ and $(7.1\pm0.9)\times10^{-9}$, respectively. The column density ratio between CH$_{3}$CHO and c-C$_{2}$H$_{4}$O is $3.4\pm0.7$. A Pearson correlation heat map reveals strong positive correlations ($r$ $>$ 0.5) between the abundances and excitation temperatures of c-C$_{2}$H$_{4}$O and CH$_{3}$CHO, suggesting a possible chemical connection between those two molecules. To investigate this further, we conducted a two-phase warm-up chemical model using the gas-grain chemical code UCLCHEM. A comparison between our derived abundances and the predictions from our chemical model and existence model demonstrates good agreement within factors of 0.73 and 0.74, respectively. We propose that c-C$_{2}$H$_{4}$O may form in G358.93$-$0.03 MM1 via the grain surface reaction between C$_{2}$H$_{4}$ and O, but CH$_{3}$CHO may be produced through the surface reaction between CH$_{3}$ and HCO.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 14:58:28 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Manna",
"Arijit",
""
],
[
"Pal",
"Sabyasachi",
""
]
] |
2509.03383
|
Yiyang Huang
|
Yiyang Huang, Zixuan Wang, Zishen Wan, Yapeng Tian, Haobo Xu, Yinhe Han, Yiming Gan
|
ANNIE: Be Careful of Your Robots
| null | null | null | null |
cs.AI cs.RO
|
http://creativecommons.org/licenses/by/4.0/
|
The integration of vision-language-action (VLA) models into embodied AI (EAI) robots is rapidly advancing their ability to perform complex, long-horizon tasks in humancentric environments. However, EAI systems introduce critical security risks: a compromised VLA model can directly translate adversarial perturbations on sensory input into unsafe physical actions. Traditional safety definitions and methodologies from the machine learning community are no longer sufficient. EAI systems raise new questions, such as what constitutes safety, how to measure it, and how to design effective attack and defense mechanisms in physically grounded, interactive settings. In this work, we present the first systematic study of adversarial safety attacks on embodied AI systems, grounded in ISO standards for human-robot interactions. We (1) formalize a principled taxonomy of safety violations (critical, dangerous, risky) based on physical constraints such as separation distance, velocity, and collision boundaries; (2) introduce ANNIEBench, a benchmark of nine safety-critical scenarios with 2,400 video-action sequences for evaluating embodied safety; and (3) ANNIE-Attack, a task-aware adversarial framework with an attack leader model that decomposes long-horizon goals into frame-level perturbations. Our evaluation across representative EAI models shows attack success rates exceeding 50% across all safety categories. We further demonstrate sparse and adaptive attack strategies and validate the real-world impact through physical robot experiments. These results expose a previously underexplored but highly consequential attack surface in embodied AI systems, highlighting the urgent need for security-driven defenses in the physical AI era. Code is available at https://github.com/RLCLab/Annie.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:00:28 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Huang",
"Yiyang",
""
],
[
"Wang",
"Zixuan",
""
],
[
"Wan",
"Zishen",
""
],
[
"Tian",
"Yapeng",
""
],
[
"Xu",
"Haobo",
""
],
[
"Han",
"Yinhe",
""
],
[
"Gan",
"Yiming",
""
]
] |
2509.03384
|
Fernando Lled\'o
|
Eva A. Gallardo-Guti\'errez, Fernando Lled\'o and Laura S\'aenz
|
Finite dimensional approximations for Hilbert space operators and applications in Quantum Mechanics
|
17 pages
| null | null | null |
math.FA math-ph math.MP math.OA
|
http://creativecommons.org/licenses/by/4.0/
|
In this work, we develop a unified framework for quasidiagonal and F\o lner-type approximations of linear operators on Hilbert spaces. These approximations (originally formulated for bounded operators and operator algebras) involve sequences of non-zero finite rank orthogonal projections that asymptotically commute with the operator -- either in norm (quasidiagonal) or in mean (F\o lner). Such structures guarantee spectral approximation results in terms of their finite sections. We extend this theory to unbounded, densely defined closable operators, establishing a generalization of Halmos' classical result: every closable quasidiagonal operator is a compact perturbation of a closable block-diagonal operator on the same domain. Likewise, we introduce sparse F\o lner sequences and establish an interplay between quasidiagonal approximations and the existence of sparse F\o lner sequences. The theoretical developments are illustrated with explicit examples using different types of weighted shifts and applied to quantum mechanical models, including a detailed treatment of the Weyl algebra and its Schr\"odinger representation.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:01:50 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Gallardo-Gutiérrez",
"Eva A.",
""
],
[
"Lledó",
"Fernando",
""
],
[
"Sáenz",
"Laura",
""
]
] |
2509.03385
|
Reina Ishikawa
|
Reina Ishikawa, Ryo Fujii, Hideo Saito and Ryo Hachiuma
|
Human Preference-Aligned Concept Customization Benchmark via Decomposed Evaluation
|
Accepted to ICCV Workshop 2025
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Evaluating concept customization is challenging, as it requires a comprehensive assessment of fidelity to generative prompts and concept images. Moreover, evaluating multiple concepts is considerably more difficult than evaluating a single concept, as it demands detailed assessment not only for each individual concept but also for the interactions among concepts. While humans can intuitively assess generated images, existing metrics often provide either overly narrow or overly generalized evaluations, resulting in misalignment with human preference. To address this, we propose Decomposed GPT Score (D-GPTScore), a novel human-aligned evaluation method that decomposes evaluation criteria into finer aspects and incorporates aspect-wise assessments using Multimodal Large Language Model (MLLM). Additionally, we release Human Preference-Aligned Concept Customization Benchmark (CC-AlignBench), a benchmark dataset containing both single- and multi-concept tasks, enabling stage-wise evaluation across a wide difficulty range -- from individual actions to multi-person interactions. Our method significantly outperforms existing approaches on this benchmark, exhibiting higher correlation with human preferences. This work establishes a new standard for evaluating concept customization and highlights key challenges for future research. The benchmark and associated materials are available at https://github.com/ReinaIshikawa/D-GPTScore.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:02:40 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Ishikawa",
"Reina",
""
],
[
"Fujii",
"Ryo",
""
],
[
"Saito",
"Hideo",
""
],
[
"Hachiuma",
"Ryo",
""
]
] |
2509.03386
|
Jia He
|
Ziye Jia, Jia He, Yuanhao Cui, Qiuming Zhu, Ligang Yuan, Fuhui Zhou, Qihui Wu, Dusit Niyato and Zhu Han
|
Hierarchical Low-Altitude Wireless Network Empowered Air Traffic Management
|
7 pages 6 figures
| null | null | null |
cs.NI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
As the increasing development of low-altitude aircrafts, the rational design of low-altitude networks directly impacts the aerial safety and resource utilization. To address the challenges of environmental complexity and aircraft diversity in the traffic management, we propose a hierarchical low-altitude wireless network (HLWN) framework. Empowered by the threedimensional spatial discretization and integrated wireless monitoring mechanisms in HLWN, we design low-altitude air corridors to guarantee safe operation and optimization. Besides, we develop the multi-dimensional flight risk assessment through conflict detection and probabilistic collision analysis, facilitating dynamic collision avoidance for heterogeneous aircrafts. Finally, the open issues and future directions are investigated to provide insights into HLAN development.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:04:42 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Jia",
"Ziye",
""
],
[
"He",
"Jia",
""
],
[
"Cui",
"Yuanhao",
""
],
[
"Zhu",
"Qiuming",
""
],
[
"Yuan",
"Ligang",
""
],
[
"Zhou",
"Fuhui",
""
],
[
"Wu",
"Qihui",
""
],
[
"Niyato",
"Dusit",
""
],
[
"Han",
"Zhu",
""
]
] |
2509.03387
|
Leonardo Bellinato Giacomelli
|
Leonardo Bellinato Giacomelli, Thomas Bland, Louis Lafforgue, Francesca Ferlaino, Manfred J. Mark and Luca Barbiero
|
Topology meets superconductivity in a one-dimensional $t-J$ model of magnetic atoms
|
13 pages, 9 figures
| null | null | null |
cond-mat.quant-gas cond-mat.str-el quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Strongly interacting fermions represent the key constituent of several intriguing phases of matter. However, due to the inherent complexity of these systems, important regimes are still inaccessible. Here, we derive a realistic and flexible setup based on ultracold magnetic lanthanide atoms trapped in a one-dimensional optical lattice. Leveraging their large magnetic moments, we design a fermionic $t-J$ model with independently tunable hopping, spin-spin couplings, and onsite interaction. Through combined analytical and numerical analysis, we uncover a variety of many-body quantum phases$-$including superconducting and topological states. Crucially, in the regime of attractive onsite interaction we reveal that topology and superconductivity coexist, thus giving rise to an exotic state of matter: a topological triplet superconductor. We also outline a practical protocol to prepare and detect all discovered phases using current experimental techniques. Our results establish an alternative and powerful route for a deeper understanding of strongly interacting fermionic quantum matter.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:05:11 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Giacomelli",
"Leonardo Bellinato",
""
],
[
"Bland",
"Thomas",
""
],
[
"Lafforgue",
"Louis",
""
],
[
"Ferlaino",
"Francesca",
""
],
[
"Mark",
"Manfred J.",
""
],
[
"Barbiero",
"Luca",
""
]
] |
2509.03388
|
Seokhoon Yun
|
Sougata Ganguly, Minxi He, Chang Sub Shin, Oscar Straniero, Seokhoon Yun
|
Probing Heavy Dark Matter in Red Giants
|
13 pages, 6 figures
| null | null | null |
hep-ph astro-ph.HE astro-ph.SR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Red giants (RGs) provide a promising astrophysical environment for capturing dark matter (DM) via elastic scattering with stellar nuclei. Captured DM particles migrate toward the helium-rich core and accumulate into a compact configuration. As the DM population grows, it can become self-gravitating and undergo gravitational collapse, leading to adiabatic contraction through interactions with the ambient medium. The resulting energy release, through elastic scattering and, where relevant, DM annihilation during collapse, locally heats the stellar core and can trigger helium ignition earlier than that predicted by standard stellar evolution. We analyze the conditions under which DM-induced heating leads to runaway helium burning and identify the critical DM mass required for ignition. Imposing the observational constraint that helium ignition must not occur before the observed luminosity at the tip of the RG branch, we translate these conditions into bounds on DM properties. Remarkably, we find that RGs are sensitive to DM, particularly with masses around $10^{11} \,{\rm GeV}$ and spin-independent scattering cross sections near $10^{-37}\,{\rm cm}^2$, which is comparable to the reach of current terrestrial direct detection experiments.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:06:45 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Ganguly",
"Sougata",
""
],
[
"He",
"Minxi",
""
],
[
"Shin",
"Chang Sub",
""
],
[
"Straniero",
"Oscar",
""
],
[
"Yun",
"Seokhoon",
""
]
] |
2509.03389
|
Shreyasi Mukherjee
|
Dario Fasone, Shreyasi Mukherjee, Dario Penna, Fabio Cirinn\`a, Mauro Paternostro, Elisabetta Paladino, Luigi Giannelli, and Giuseppe A. Falci
|
Detection of noise correlations in two qubit systems by Machine Learning
|
11 pages, 7 figures
| null | null | null |
quant-ph cond-mat.other
|
http://creativecommons.org/licenses/by/4.0/
|
We introduce and validate a machine-learning assisted quantum sensing protocol to classify spatial and temporal correlations of classical noise affecting two ultrastrongly coupled qubits. We consider six distinct classes of Markovian and non-Markovian noise. Leveraging the sensitivity of a coherent population transfer protocol under three distinct driving conditions, the various forms of noise are discriminated by only measuring the final transfer efficiencies. Our approach achieves $\gtrsim 86\%$ accuracy in classification providing a near-perfect discrimination between Markovian and non-Markovian noise. The method requires minimal experimental resources, relying on a simple driving scheme providing three inputs to a shallow neural network with no need of measuring time-series data or real-time monitoring. The machine-learning data analysis acquires information from non-idealities of the coherent protocol highlighting how combining these techniques may significantly improve the characterization of quantum-hardware.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:13:01 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Fasone",
"Dario",
""
],
[
"Mukherjee",
"Shreyasi",
""
],
[
"Penna",
"Dario",
""
],
[
"Cirinnà",
"Fabio",
""
],
[
"Paternostro",
"Mauro",
""
],
[
"Paladino",
"Elisabetta",
""
],
[
"Giannelli",
"Luigi",
""
],
[
"Falci",
"Giuseppe A.",
""
]
] |
2509.03390
|
Peter Mur\v{s}i\v{c}
|
Eric Gottlieb, Dawood Khatana, Matja\v{z} Krnc, Peter Mur\v{s}i\v{c}, Ismael Qureshi
|
Row Impartial Terminus
| null | null | null | null |
math.CO cs.DM
|
http://creativecommons.org/publicdomain/zero/1.0/
|
We introduce Row Impartial Terminus (RIT), an impartial combinatorial game played on integer partitions. We show that any position in RIT can be uniquely decomposed into a core and a remnant. Our central result is that the Conway pair of any RIT position-which determines the outcome under both normal and mis\`ere play-is identical to the Conway pair of a corresponding position in the game of Nim defined by the remnant. This finding provides a complete winning strategy for both variants of RIT, reducing its analysis to the well-understood framework of Nim. As a consequence, we classify RIT within the Conway-Gurvich-Ho hierarchy, showing it to be forced and miserable but not pet.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:14:03 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Gottlieb",
"Eric",
""
],
[
"Khatana",
"Dawood",
""
],
[
"Krnc",
"Matjaž",
""
],
[
"Muršič",
"Peter",
""
],
[
"Qureshi",
"Ismael",
""
]
] |
2509.03391
|
Trent Buskirk
|
Trent D. Buskirk, Florian Keusch, Leah von der Heyde and Adam Eck
|
More Parameters Than Populations: A Systematic Literature Review of Large Language Models within Survey Research
| null | null | null | null |
cs.DL cs.CY
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Survey research has a long-standing history of being a human-powered field, but one that embraces various technologies for the collection, processing, and analysis of various behavioral, political, and social outcomes of interest, among others. At the same time, Large Language Models (LLMs) bring new technological challenges and prerequisites in order to fully harness their potential. In this paper, we report work-in-progress on a systematic literature review based on keyword searches from multiple large-scale databases as well as citation networks that assesses how LLMs are currently being applied within the survey research process. We synthesize and organize our findings according to the survey research process to include examples of LLM usage across three broad phases: pre-data collection, data collection, and post-data collection. We discuss selected examples of potential use cases for LLMs as well as its pitfalls based on examples from existing literature. Considering survey research has rich experience and history regarding data quality, we discuss some opportunities and describe future outlooks for survey research to contribute to the continued development and refinement of LLMs.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:15:31 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Buskirk",
"Trent D.",
""
],
[
"Keusch",
"Florian",
""
],
[
"von der Heyde",
"Leah",
""
],
[
"Eck",
"Adam",
""
]
] |
2509.03392
|
Xinyue Chen
|
Xinyue Chen, Kunlin Ruan, Kexin Phyllis Ju, Nathan Yap, Xu Wang
|
More AI Assistance Reduces Cognitive Engagement: Examining the AI Assistance Dilemma in AI-Supported Note-Taking
|
Accepted by CSCW2025
| null |
10.1145/3757632
| null |
cs.HC
|
http://creativecommons.org/licenses/by/4.0/
|
As AI tools become increasingly embedded in cognitively demanding tasks such as note-taking, questions remain about whether they enhance or undermine cognitive engagement. This paper examines the "AI Assistance Dilemma" in note-taking, investigating how varying levels of AI support affect user engagement and comprehension. In a within-subject experiment, we asked participants (N=30) to take notes during lecture videos under three conditions: Automated AI (high assistance with structured notes), Intermediate AI (moderate assistance with real-time summary, and Minimal AI (low assistance with transcript). Results reveal that Intermediate AI yields the highest post-test scores and Automated AI the lowest. Participants, however, preferred the automated setup due to its perceived ease of use and lower cognitive effort, suggesting a discrepancy between preferred convenience and cognitive benefits. Our study provides insights into designing AI assistance that preserves cognitive engagement, offering implications for designing moderate AI support in cognitive tasks.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:15:39 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Chen",
"Xinyue",
""
],
[
"Ruan",
"Kunlin",
""
],
[
"Ju",
"Kexin Phyllis",
""
],
[
"Yap",
"Nathan",
""
],
[
"Wang",
"Xu",
""
]
] |
2509.03393
|
Taisiya Khakharova
|
Taisiya Khakharova, Lucas Sakizloglou, Leen Lambers
|
Exploring a Graph-based Approach to Offline Reinforcement Learning for Sepsis Treatment
|
18th European Workshop on Reinforcement Learning (EWRL 2025)
| null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Sepsis is a serious, life-threatening condition. When treating sepsis, it is challenging to determine the correct amount of intravenous fluids and vasopressors for a given patient. While automated reinforcement learning (RL)-based methods have been used to support these decisions with promising results, previous studies have relied on relational data. Given the complexity of modern healthcare data, representing data as a graph may provide a more natural and effective approach. This study models patient data from the well-known MIMIC-III dataset as a heterogeneous graph that evolves over time. Subsequently, we explore two Graph Neural Network architectures - GraphSAGE and GATv2 - for learning patient state representations, adopting the approach of decoupling representation learning from policy learning. The encoders are trained to produce latent state representations, jointly with decoders that predict the next patient state. These representations are then used for policy learning with the dBCQ algorithm. The results of our experimental evaluation confirm the potential of a graph-based approach, while highlighting the complexity of representation learning in this domain.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:15:41 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Khakharova",
"Taisiya",
""
],
[
"Sakizloglou",
"Lucas",
""
],
[
"Lambers",
"Leen",
""
]
] |
2509.03394
|
Amirhossein Shahbazinia
|
Amirhossein Shahbazinia, Darong Huang, Luis Costero, David Atienza
|
CloudFormer: An Attention-based Performance Prediction for Public Clouds with Unknown Workload
| null | null | null | null |
cs.DC cs.LG cs.PF
|
http://creativecommons.org/licenses/by/4.0/
|
Cloud platforms are increasingly relied upon to host diverse, resource-intensive workloads due to their scalability, flexibility, and cost-efficiency. In multi-tenant cloud environments, virtual machines are consolidated on shared physical servers to improve resource utilization. While virtualization guarantees resource partitioning for CPU, memory, and storage, it cannot ensure performance isolation. Competition for shared resources such as last-level cache, memory bandwidth, and network interfaces often leads to severe performance degradation. Existing management techniques, including VM scheduling and resource provisioning, require accurate performance prediction to mitigate interference. However, this remains challenging in public clouds due to the black-box nature of VMs and the highly dynamic nature of workloads. To address these limitations, we propose CloudFormer, a dual-branch Transformer-based model designed to predict VM performance degradation in black-box environments. CloudFormer jointly models temporal dynamics and system-level interactions, leveraging 206 system metrics at one-second resolution across both static and dynamic scenarios. This design enables the model to capture transient interference effects and adapt to varying workload conditions without scenario-specific tuning. Complementing the methodology, we provide a fine-grained dataset that significantly expands the temporal resolution and metric diversity compared to existing benchmarks. Experimental results demonstrate that CloudFormer consistently outperforms state-of-the-art baselines across multiple evaluation metrics, achieving robust generalization across diverse and previously unseen workloads. Notably, CloudFormer attains a mean absolute error (MAE) of just 7.8%, representing a substantial improvement in predictive accuracy and outperforming existing methods at least by 28%.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:15:44 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Shahbazinia",
"Amirhossein",
""
],
[
"Huang",
"Darong",
""
],
[
"Costero",
"Luis",
""
],
[
"Atienza",
"David",
""
]
] |
2509.03395
|
Emma Wallace
|
Emma Wallace, Christian Iliadis, Sumner Starrfield
|
Elemental and Isotopic Yields from T Coronae Borealis: Predictions and Uncertainties
|
13 pages, 3 figures, 3 tables. Accepted for publication in The Astrophysical Journal
| null | null | null |
astro-ph.SR astro-ph.HE
|
http://creativecommons.org/licenses/by-sa/4.0/
|
T Coronae Borealis (T CrB) is a symbiotic recurrent nova system expected to undergo its next outburst within the next two years. Recent hydrodynamic simulations have predicted the nucleosynthetic yields for both carbon-oxygen (CO) and oxygen-neon (ONe) white-dwarf models, but without accounting for thermonuclear reaction-rate uncertainties. We perform detailed Monte Carlo post-processing nucleosynthesis calculations based on updated thermonuclear reaction rates and uncertainties from the 2025 evaluation. We quantify the resulting abundance uncertainties and identify the key nuclear reactions that dominate them. Our results show that both the CO and ONe nova models robustly produce characteristic CNO isotopes. More pronounced abundance differences emerge for elements with A $\ge$ 20. Sulfur is the most robust observational discriminator between the CO and ONe nova models, with a model-to-model difference of a factor of $\approx$30 and minimal sensitivity to reaction rate uncertainties. Neon, silicon, and phosphorus exhibit even larger abundance differences (factors of $\approx$150-250), providing strong diagnostic potential. While their predicted yields are subject to larger uncertainties, these remain smaller than the model-to-model differences, allowing these elements to serve as useful, though less precise, tracers of white-dwarf composition. Chlorine, argon, and potassium also differ between models, but the 1$\sigma$-abundance ranges for the CO and ONe models overlap, reducing their present usefulness as composition tracers. We find that only nine nuclear reactions dominate the abundance uncertainties of the most diagnostically important isotopes, and their influence is largely independent of the underlying white-dwarf composition. These results provide guidance for future experimental efforts and for interpreting ejecta compositions in the next eruption of T CrB.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:15:54 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Wallace",
"Emma",
""
],
[
"Iliadis",
"Christian",
""
],
[
"Starrfield",
"Sumner",
""
]
] |
2509.03396
|
Advika Rajapakse
|
Advika Rajapakse
|
On Steenrod squares for even and odd Khovanov homology
|
88 pages
| null | null | null |
math.GT math.AT
|
http://creativecommons.org/licenses/by/4.0/
|
For an arbitrary link $L \subset S^3$ , Sarkar-Scaduto-Stoffregen construct a family of spatial refinements of even and odd Khovanov homology. We give a computation of $\text{Sq}^2$ on these spaces, determining their stable homotopy types for all knots K up to 11 crossings. We also prove that the Steenrod squares $\text{Sq}_0^2$ , $\text{Sq}_1^2$ defined by Sch\"utz do arise as Steenrod squares on these spaces.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:17:19 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Rajapakse",
"Advika",
""
]
] |
2509.03398
|
Majid Aalizadeh
|
Majid Aalizadeh, Chinmay Raut, Ali Tabartehfarahani, Xudong Fan
|
Machine Learning-Enhanced Colorimetric Sensing: Achieving over 5700-fold Accuracy Improvement via Full-Spectrum Modeling
|
24 pages, 7 figures, 1 table
| null | null | null |
physics.med-ph math-ph math.MP q-bio.QM
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Conventional colorimetric sensing methods typically rely on signal intensity at a single wavelength, often selected heuristically based on peak visual modulation. This approach overlooks the structured information embedded in full-spectrum transmission profiles, particularly in intensity-based systems where linear models may be highly effective. In this study, we experimentally demonstrate that applying a forward feature selection strategy to normalized transmission spectra, combined with linear regression and ten-fold cross-validation, yields significant improvements in predictive accuracy. Using food dye dilutions as a model system, the mean squared error was reduced from over 22,000 with a single wavelength to 3.87 using twelve selected features, corresponding to a more than 5,700-fold enhancement. These results validate that full-spectrum modeling enables precise concentration prediction without requiring changes to the sensing hardware. The approach is broadly applicable to colorimetric assays used in medical diagnostics, environmental monitoring, and industrial analysis, offering a scalable pathway to improve sensitivity and reliability in existing platforms.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:20:36 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Aalizadeh",
"Majid",
""
],
[
"Raut",
"Chinmay",
""
],
[
"Tabartehfarahani",
"Ali",
""
],
[
"Fan",
"Xudong",
""
]
] |
2509.03399
|
Marcel Blattner
|
Marcel Blattner
|
Tangential Action Spaces: Geometry, Memory and Cost in Holonomic and Nonholonomic Agents
|
28 pages, 6 figures
| null | null | null |
eess.SY cs.SY
|
http://creativecommons.org/licenses/by/4.0/
|
How much energy must an embodied agent spend to remember its past actions? We present Tangential Action Spaces (TAS), a differential-geometric framework revealing a fundamental trade-off between memory and energy in embodied agents. By modeling agents as hierarchical manifolds with projections Phi: P -> C and Psi: C -> I connecting physical (P), cognitive (C), and intentional (I) spaces, we show that the geometry of Phi dictates both memory mechanisms and their energetic costs. Our main contributions are: (1) a rigorous classification proving that one-to-one projections (diffeomorphisms) require engineered dynamics for memory while many-to-one projections (fibrations) enable intrinsic geometric memory through connection curvature; (2) a proof that any deviation from the energy-minimal lift incurs a quantifiable penalty, establishing that path-dependent behavior necessarily costs energy; and (3) a universal principle that excess cost Delta E scales with the square of accumulated holonomy (geometric memory). We validate this cost-memory duality through five systems: the strip-sine system (engineered memory, Delta E proportional to (Delta h)^2), helical and twisted fibrations (intrinsic geometric memory), and flat/cylindrical fibrations (proving curvature, not topology, creates memory). This framework bridges geometric mechanics and embodied cognition, explaining biological motor diversity and providing design principles for efficient robotic control.
|
[
{
"version": "v1",
"created": "Tue, 02 Sep 2025 15:36:22 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Blattner",
"Marcel",
""
]
] |
2509.03400
|
Krzysztof Marek Kutak
|
Martin Hentschinski, Hannes Jung, Krzysztof Kutak
|
Entanglement entropy, Monte Carlo event generators, and soft gluons DIScovery
|
17 pages
| null | null | null |
hep-ph hep-th quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study entropy production in Deep Inelastic Scattering using Monte Carlo simulations. We show that the dominant contribution to entropy is due to soft gluons. This contribution is usually neglected in standard Monte Carlo approaches, since it does not affect hadronic spectra. However, it is relevant for entropy and multiplicity distributions, as we demonstrate with explicit calculations. We further show that, as one lowers the cutoff - i.e. includes soft gluons - thereby making the Monte Carlo parton distributions closer to inclusive PDFs, the resulting entropy is very close to that measured by the H1 collaboration, without the need to account for hadronization effects. This provides further evidence that the bulk of the measured entropy originates from initial-state effects.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:24:04 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Hentschinski",
"Martin",
""
],
[
"Jung",
"Hannes",
""
],
[
"Kutak",
"Krzysztof",
""
]
] |
2509.03401
|
MD Zaibul Anam
|
Md Zaibul Anam, Ogheneyoma Aghoghovbia, Mohammed Al-Fahdi, Lingyu Kong, Victor Fung, and Ming Hu
|
A Comprehensive Assessment and Benchmark Study of Large Atomistic Foundation Models for Phonons
| null | null | null | null |
cond-mat.mtrl-sci
|
http://creativecommons.org/licenses/by/4.0/
|
The rapid development of universal machine learning potentials (uMLPs) has enabled efficient, accurate predictions of diverse material properties across broad chemical spaces. While their capability for modeling phonon properties is emerging, systematic benchmarking across chemically diverse systems remains limited. We evaluate six recent uMLPs (EquiformerV2, MatterSim, MACE, and CHGNet) on 2,429 crystalline materials from the Open Quantum Materials Database. Models were used to compute atomic forces in displaced supercells, derive interatomic force constants (IFCs), and predict phonon properties including lattice thermal conductivity (LTC), compared with density functional theory (DFT) and experimental data. The EquiformerV2 pretrained model trained on the OMat24 dataset exhibits strong performance in predicting atomic forces and third-order IFC, while its fine-tuned counterpart consistently outperforms other models in predicting second-order IFC, LTC, and other phonon properties. Although MACE and CHGNet demonstrated comparable force prediction accuracy to EquiformerV2, notable discrepancies in IFC fitting led to poor LTC predictions. Conversely, MatterSim, despite lower force accuracy, achieved intermediate IFC predictions, suggesting error cancellation and complex relationships between force accuracy and phonon predictions. This benchmark guides the evaluation and selection of uMLPs for high-throughput screening of materials with targeted thermal transport properties.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:24:45 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Anam",
"Md Zaibul",
""
],
[
"Aghoghovbia",
"Ogheneyoma",
""
],
[
"Al-Fahdi",
"Mohammed",
""
],
[
"Kong",
"Lingyu",
""
],
[
"Fung",
"Victor",
""
],
[
"Hu",
"Ming",
""
]
] |
2509.03402
|
Zofia Budzik
|
Zofia Budzik and Matylda {\L}ukaszewicz and Rados{\l}aw Poleski
|
Metallicity of the double Red Clump in the Milky Way Bulge
|
ApJ accepted
| null | null | null |
astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
We present a reanalysis of the metallicity of the double red clump (RC) stars from the Milky Way bulge. Two leading explanations for the existence of the double RC concern possible differences in chemical composition (multiple populations of stars) or distance (X-shaped bulge). We aim to verify the chemical composition hypothesis by determining the mean metallicity of each RC. We use infrared photometric data and metallicities from a previous study (that are based on the Michigan/Magellan Fiber System spectra). In contrast with previous studies, we assign stars to RCs or the red giant branch using a Bayesian approach. Our resulting difference between mean metallicity of the two RCs equals: $0.23^{+0.19}_{-0.24}$ or $0.28^{+0.17}_{-0.18}~\mathrm{dex}$, if the red giant branch bump is included or not, respectively. Because of the high statistical error, the result does not confirm the multiple populations hypothesis.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:27:36 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Budzik",
"Zofia",
""
],
[
"Łukaszewicz",
"Matylda",
""
],
[
"Poleski",
"Radosław",
""
]
] |
2509.03403
|
Chenlu Ye
|
Chenlu Ye, Zhou Yu, Ziji Zhang, Hao Chen, Narayanan Sadagopan, Jing Huang, Tong Zhang, Anurag Beniwal
|
Beyond Correctness: Harmonizing Process and Outcome Rewards through RL Training
| null | null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Reinforcement learning with verifiable rewards (RLVR) has emerged to be a predominant paradigm for mathematical reasoning tasks, offering stable improvements in reasoning ability. However, Outcome Reward Models (ORMs) in RLVR are too coarse-grained to distinguish flawed reasoning within correct answers or valid reasoning within incorrect answers. This lack of granularity introduces noisy and misleading gradients significantly and hinders further progress in reasoning process quality. While Process Reward Models (PRMs) offer fine-grained guidance for intermediate steps, they frequently suffer from inaccuracies and are susceptible to reward hacking.
To resolve this dilemma, we introduce PRocess cOnsistency Filter (PROF), an effective data process curation method that harmonizes noisy, fine-grained process rewards with accurate, coarse-grained outcome rewards. Rather than naively blending PRM and ORM in the objective function (arXiv:archive/2506.18896), PROF leverages their complementary strengths through consistency-driven sample selection. Our approach retains correct responses with higher averaged process values and incorrect responses with lower averaged process values, while maintaining positive/negative training sample balance. Extensive experiments demonstrate that our method not only consistently improves the final accuracy over $4\%$ compared to the blending approaches, but also strengthens the quality of intermediate reasoning steps. Codes and training recipes are available at https://github.com/Chenluye99/PROF.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:28:51 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Ye",
"Chenlu",
""
],
[
"Yu",
"Zhou",
""
],
[
"Zhang",
"Ziji",
""
],
[
"Chen",
"Hao",
""
],
[
"Sadagopan",
"Narayanan",
""
],
[
"Huang",
"Jing",
""
],
[
"Zhang",
"Tong",
""
],
[
"Beniwal",
"Anurag",
""
]
] |
2509.03404
|
Francisco Arizo-Borillo
|
F. D. Arizo-Borillo (Centro de Estudios de Fisica del Cosmos de Aragon, CEFCA), C. Lopez-Sanjuan (CEFCA, Unidad Asociada CEFCA-IAA), I. Pintos-Castro (CEFCA), J. A. Fernandez-Ontiveros (CEFCA, Unidad Asociada CEFCA-IAA), T. Kuutma (CEFCA), A. Lumbreras-Calle (CEFCA), A. Hernan-Caballero (CEFCA, Unidad Asociada CEFCA-IAA), H. Dominguez-Sanchez (Instituto de Fisica de Cantabria-CSIC, CEFCA), G. De Lucia (INAF - Astronomical Observatory of Trieste, IFPU - Institute for Fundamental Physics of the Universe), F. Fontanot (INAF - Astronomical Observatory of Trieste, IFPU - Institute for Fundamental Physics of the Universe), L. A. Diaz-Garcia (Instituto de Astrofisica de Andalucia, CSIC), J. M. Vilchez (Instituto de Astrofisica de Andalucia, CSIC), P. T. Rahna (CEFCA), A. J. Cenarro (CEFCA, Unidad Asociada CEFCA-IAA), D. Cristobal-Hornillos (CEFCA), C. Hernandez-Monteagudo (Instituto de Astrofisica de Canarias, Universidad de La Laguna), A. Marin-Franch (CEFCA, Unidad Asociada CEFCA-IAA), M. Moles (CEFCA), J. Varela (CEFCA), H. Vazquez Ramio (CEFCA, Unidad Asociada CEFCA-IAA), J. Alcaniz (Observatorio Nacional - MCTI, Brazil), R. A. Dupke (Observatorio Nacional - MCTI, Brazil, University of Michigan), A. Ederoclite (CEFCA, Unidad Asociada CEFCA-IAA), L. Sodre Jr. (Universidade de Sao Paulo), R. E. Angulo (Donostia International Physics Centre, DIPC, IKERBASQUE - Basque Foundation for Science)
|
J-PLUS: The stellar mass function of quiescent and star forming galaxies at 0.05 <= z <= 0.2
|
14 pages, 12 figures, 3 tables. Submitted to Astronomy & Astrophysics
| null | null | null |
astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
Aims. We derive the stellar mass function (SMF) of quiescent and star-forming galaxies at z <= 0.2 using 12-band optical photometry from the third data release (DR3) of the Javalambre Photometric Local Universe Survey (J-PLUS) over 3,284 deg^2. Methods. We select approximately 890,000 galaxies with r <= 20 mag and photometric redshifts in the range 0.05 <= z <= 0.20. Stellar masses and star formation rates were derived through spectral energy distribution fitting with CIGALE, confronted with spectroscopic samples. Galaxies are classified as star-forming or quiescent based on their specific star formation rate (sSFR), adopting log(sSFR [yr^-1]) = -10.2. We compute SMFs for both populations using the 1/Vmax method, apply completeness corrections, and fit Schechter functions. Results. The SMFs from J-PLUS DR3 are well described by Schechter functions and agree with previous photometric and spectroscopic studies. The characteristic mass for quiescent galaxies, log(M*/Msun) = 10.80, is 0.4 dex larger than that of star-forming galaxies. The faint-end slope is steeper for star-forming galaxies (alpha = -1.2) than for quiescent ones (alpha = -0.7). The quiescent fraction rises by 40 percent per dex in stellar mass, reaching fQ > 0.95 at log(M*/Msun) > 11. Comparisons with the GAEA semi-analytic model reveal an excess of star-forming galaxies at intermediate masses. Conclusions. J-PLUS DR3 stellar mass functions and quiescent fractions are consistent with the literature and provide robust constraints for galaxy formation models. Quiescent galaxies represent 45 percent of number density above log(M*) > 9, but 75 percent of stellar mass density. The use of 12 optical bands, including 7 narrow filters, improves redshift precision by 20 percent, enabling more accurate SED fitting and galaxy classification.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:29:08 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Arizo-Borillo",
"F. D.",
"",
"Centro de Estudios de Fisica del Cosmos de Aragon, CEFCA"
],
[
"Lopez-Sanjuan",
"C.",
"",
"CEFCA, Unidad Asociada CEFCA-IAA"
],
[
"Pintos-Castro",
"I.",
"",
"CEFCA"
],
[
"Fernandez-Ontiveros",
"J. A.",
"",
"CEFCA, Unidad Asociada CEFCA-IAA"
],
[
"Kuutma",
"T.",
"",
"CEFCA"
],
[
"Lumbreras-Calle",
"A.",
"",
"CEFCA"
],
[
"Hernan-Caballero",
"A.",
"",
"CEFCA, Unidad Asociada CEFCA-IAA"
],
[
"Dominguez-Sanchez",
"H.",
"",
"Instituto de Fisica de Cantabria-CSIC, CEFCA"
],
[
"De Lucia",
"G.",
"",
"INAF - Astronomical Observatory of Trieste, IFPU - Institute for Fundamental Physics of the Universe"
],
[
"Fontanot",
"F.",
"",
"INAF - Astronomical Observatory of Trieste, IFPU - Institute for Fundamental Physics of the Universe"
],
[
"Diaz-Garcia",
"L. A.",
"",
"Instituto de Astrofisica de Andalucia, CSIC"
],
[
"Vilchez",
"J. M.",
"",
"Instituto de Astrofisica de Andalucia, CSIC"
],
[
"Rahna",
"P. T.",
"",
"CEFCA"
],
[
"Cenarro",
"A. J.",
"",
"CEFCA, Unidad Asociada CEFCA-IAA"
],
[
"Cristobal-Hornillos",
"D.",
"",
"CEFCA"
],
[
"Hernandez-Monteagudo",
"C.",
"",
"Instituto de Astrofisica de Canarias, Universidad de La Laguna"
],
[
"Marin-Franch",
"A.",
"",
"CEFCA, Unidad Asociada CEFCA-IAA"
],
[
"Moles",
"M.",
"",
"CEFCA"
],
[
"Varela",
"J.",
"",
"CEFCA"
],
[
"Ramio",
"H. Vazquez",
"",
"CEFCA, Unidad Asociada CEFCA-IAA"
],
[
"Alcaniz",
"J.",
"",
"Observatorio Nacional - MCTI, Brazil"
],
[
"Dupke",
"R. A.",
"",
"Observatorio Nacional - MCTI, Brazil, University of Michigan"
],
[
"Ederoclite",
"A.",
"",
"CEFCA, Unidad Asociada CEFCA-IAA"
],
[
"Sodre",
"L.",
"Jr.",
"Universidade de Sao Paulo"
],
[
"Angulo",
"R. E.",
"",
"Donostia International Physics Centre, DIPC, IKERBASQUE - Basque Foundation for Science"
]
] |
2509.03405
|
Daniela Gottesman
|
Daniela Gottesman, Alon Gilae-Dotan, Ido Cohen, Yoav Gur-Arieh, Marius Mosbach, Ori Yoran, Mor Geva
|
LMEnt: A Suite for Analyzing Knowledge in Language Models from Pretraining Data to Representations
|
Submitted to TACL, August 2025
| null | null | null |
cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Language models (LMs) increasingly drive real-world applications that require world knowledge. However, the internal processes through which models turn data into representations of knowledge and beliefs about the world, are poorly understood. Insights into these processes could pave the way for developing LMs with knowledge representations that are more consistent, robust, and complete. To facilitate studying these questions, we present LMEnt, a suite for analyzing knowledge acquisition in LMs during pretraining. LMEnt introduces: (1) a knowledge-rich pretraining corpus, fully annotated with entity mentions, based on Wikipedia, (2) an entity-based retrieval method over pretraining data that outperforms previous approaches by as much as 80.4%, and (3) 12 pretrained models with up to 1B parameters and 4K intermediate checkpoints, with comparable performance to popular open-sourced models on knowledge benchmarks. Together, these resources provide a controlled environment for analyzing connections between entity mentions in pretraining and downstream performance, and the effects of causal interventions in pretraining data. We show the utility of LMEnt by studying knowledge acquisition across checkpoints, finding that fact frequency is key, but does not fully explain learning trends. We release LMEnt to support studies of knowledge in LMs, including knowledge representations, plasticity, editing, attribution, and learning dynamics.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:31:18 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Gottesman",
"Daniela",
""
],
[
"Gilae-Dotan",
"Alon",
""
],
[
"Cohen",
"Ido",
""
],
[
"Gur-Arieh",
"Yoav",
""
],
[
"Mosbach",
"Marius",
""
],
[
"Yoran",
"Ori",
""
],
[
"Geva",
"Mor",
""
]
] |
2509.03406
|
Guojun Wei
|
Guojun Wei, Manzi Nan, Pengcheng Li, Yongjia Wang, Qingfeng Li, Gaochan Yong, Fuhu Liu
|
Bayesian analysis of properties of nuclear matter with the FOPI experimental data
|
10 pages, 7 figures
| null | null | null |
nucl-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Based on the ultra-relativistic quantum molecular dynamics (UrQMD) transport model, combined with experimental data of directed flow, elliptic flow, and nuclear stopping power measured by FOPI in $\rm ^{197}Au+^{197}Au$ collisions at beam energies ($E_{lab}$) of 0.25 and 0.4 GeV/nucleon, the incompressibility of the nuclear equation of state $K_0$, the nucleon effective mass $m^*$, and the in-medium correction factor ($F$, with respect to free-space values) on the nucleon-nucleon elastic cross sections are studied by Bayesian analysis. It is found that both $m^*$ and $F$ can be tightly constrained with the uncertainty $\le$ 15\%, however, $K_0$ cannot be constrained tightly. We deduce $m^*/m_0 = 0.78^{+0.09}_{-0.10}$ and $F = 0.75^{+0.08}_{-0.07}$ with experimental data at $E_{lab}$ = 0.25 GeV/nucleon, and the obtained values increased to $m^*/m_0 = 0.88^{+0.03}_{-0.03}$ and $F = 0.88^{+0.06}_{-0.07}$ at $E_{lab}$ = 0.4 GeV/nucleon. The obtained results are further verified with rapidity-dependent flow data.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:31:35 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Wei",
"Guojun",
""
],
[
"Nan",
"Manzi",
""
],
[
"Li",
"Pengcheng",
""
],
[
"Wang",
"Yongjia",
""
],
[
"Li",
"Qingfeng",
""
],
[
"Yong",
"Gaochan",
""
],
[
"Liu",
"Fuhu",
""
]
] |
2509.03407
|
Ido Kanter
|
Yarden Tzach, Ronit D. Gross, Ella Koresh, Shalom Rosner, Or Shpringer, Tal Halevi and Ido Kanter
|
Learning Mechanism Underlying NLP Pre-Training and Fine-Tuning
|
46 pages, 18 figures, 10 tables
| null | null | null |
cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Natural language processing (NLP) enables the understanding and generation of meaningful human language, typically using a pre-trained complex architecture on a large dataset to learn the language and next fine-tune its weights to implement a specific task. Twofold goals are examined; to understand the mechanism underlying successful pre-training and to determine the interplay between the pre-training accuracy and the fine-tuning of classification tasks. The following main results were obtained; the accuracy per token (APT) increased with its appearance frequency in the dataset, and its average over all tokens served as an order parameter to quantify pre-training success, which increased along the transformer blocks. Pre-training broke the symmetry among tokens and grouped them into finite, small, strong match token clusters, as inferred from the presented token confusion matrix. This feature was sharpened along the transformer blocks toward the output layer, enhancing its performance considerably compared with that of the embedding layer. Consequently, higher-order language structures were generated by pre-training, even though the learning cost function was directed solely at identifying a single token. These pre-training findings were reflected by the improved fine-tuning accuracy along the transformer blocks. Additionally, the output label prediction confidence was found to be independent of the average input APT, as the input meaning was preserved since the tokens are replaced primarily by strong match tokens. Finally, although pre-training is commonly absent in image classification tasks, its underlying mechanism is similar to that used in fine-tuning NLP classification tasks, hinting at its universality. The results were based on the BERT-6 architecture pre-trained on the Wikipedia dataset and fine-tuned on the FewRel and DBpedia classification tasks.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:32:50 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Tzach",
"Yarden",
""
],
[
"Gross",
"Ronit D.",
""
],
[
"Koresh",
"Ella",
""
],
[
"Rosner",
"Shalom",
""
],
[
"Shpringer",
"Or",
""
],
[
"Halevi",
"Tal",
""
],
[
"Kanter",
"Ido",
""
]
] |
2509.03408
|
Mohammed Amer
|
Mohammed Amer, Mohamed A. Suliman, Tu Bui, Nuria Garcia, Serban Georgescu
|
Scalable and Loosely-Coupled Multimodal Deep Learning for Breast Cancer Subtyping
| null | null | null | null |
cs.CV cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Healthcare applications are inherently multimodal, benefiting greatly from the integration of diverse data sources. However, the modalities available in clinical settings can vary across different locations and patients. A key area that stands to gain from multimodal integration is breast cancer molecular subtyping, an important clinical task that can facilitate personalized treatment and improve patient prognosis. In this work, we propose a scalable and loosely-coupled multimodal framework that seamlessly integrates data from various modalities, including copy number variation (CNV), clinical records, and histopathology images, to enhance breast cancer subtyping. While our primary focus is on breast cancer, our framework is designed to easily accommodate additional modalities, offering the flexibility to scale up or down with minimal overhead without requiring re-training of existing modalities, making it applicable to other types of cancers as well. We introduce a dual-based representation for whole slide images (WSIs), combining traditional image-based and graph-based WSI representations. This novel dual approach results in significant performance improvements. Moreover, we present a new multimodal fusion strategy, demonstrating its ability to enhance performance across a range of multimodal conditions. Our comprehensive results show that integrating our dual-based WSI representation with CNV and clinical health records, along with our pipeline and fusion strategy, outperforms state-of-the-art methods in breast cancer subtyping.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:33:04 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Amer",
"Mohammed",
""
],
[
"Suliman",
"Mohamed A.",
""
],
[
"Bui",
"Tu",
""
],
[
"Garcia",
"Nuria",
""
],
[
"Georgescu",
"Serban",
""
]
] |
2509.03409
|
Damien Lolive
|
Hoan My Tran and Damien Lolive and Aghilas Sini and Arnaud Delhay and Pierre-Fran\c{c}ois Marteau and David Guennec
|
Multi-level SSL Feature Gating for Audio Deepfake Detection
|
This paper has been accepted by ACM MM 2025
| null | null | null |
cs.SD cs.AI cs.MM
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Recent advancements in generative AI, particularly in speech synthesis, have enabled the generation of highly natural-sounding synthetic speech that closely mimics human voices. While these innovations hold promise for applications like assistive technologies, they also pose significant risks, including misuse for fraudulent activities, identity theft, and security threats. Current research on spoofing detection countermeasures remains limited by generalization to unseen deepfake attacks and languages. To address this, we propose a gating mechanism extracting relevant feature from the speech foundation XLS-R model as a front-end feature extractor. For downstream back-end classifier, we employ Multi-kernel gated Convolution (MultiConv) to capture both local and global speech artifacts. Additionally, we introduce Centered Kernel Alignment (CKA) as a similarity metric to enforce diversity in learned features across different MultiConv layers. By integrating CKA with our gating mechanism, we hypothesize that each component helps improving the learning of distinct synthetic speech patterns. Experimental results demonstrate that our approach achieves state-of-the-art performance on in-domain benchmarks while generalizing robustly to out-of-domain datasets, including multilingual speech samples. This underscores its potential as a versatile solution for detecting evolving speech deepfake threats.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:37:52 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Tran",
"Hoan My",
""
],
[
"Lolive",
"Damien",
""
],
[
"Sini",
"Aghilas",
""
],
[
"Delhay",
"Arnaud",
""
],
[
"Marteau",
"Pierre-François",
""
],
[
"Guennec",
"David",
""
]
] |
2509.03410
|
Yen-Chi Chen
|
Yanjiao Yang, Yen-Chi Chen
|
Markov Missing Graph: A Graphical Approach for Missing Data Imputation
|
43 pages, 7 figures
| null | null | null |
stat.ME math.ST stat.ML stat.TH
|
http://creativecommons.org/licenses/by/4.0/
|
We introduce the Markov missing graph (MMG), a novel framework that imputes missing data based on undirected graphs. MMG leverages conditional independence relationships to locally decompose the imputation model. To establish the identification, we introduce the Principle of Available Information (PAI), which guides the use of all relevant observed data. We then propose a flexible statistical learning paradigm, MMG Imputation Risk Minimization under PAI, that frames the imputation task as an empirical risk minimization problem. This framework is adaptable to various modeling choices. We develop theories of MMG, including the connection between MMG and Little's complete-case missing value assumption, recovery under missing completely at random, efficiency theory, and graph-related properties. We show the validity of our method with simulation studies and illustrate its application with a real-world Alzheimer's data set.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:38:36 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Yang",
"Yanjiao",
""
],
[
"Chen",
"Yen-Chi",
""
]
] |
2509.03411
|
Samu\"el Borza
|
Michael Albert, Samu\"el Borza, Maria Gordina
|
Geodesics on Grushin spaces
|
42 pages, 10 figures
| null | null | null |
math.DG math.MG
|
http://creativecommons.org/licenses/by/4.0/
|
We consider higher-dimensional generalizations of the $\alpha$-Grushin plane, focusing on the problem of classification of geodesics that minimize length, also known as optimal synthesis. Solving Hamilton's equations on these spaces using the calculus of generalized trigonometric functions, we obtain explicit conjugate times for geodesics starting at a Riemannian point. From symmetries in the geodesic structure, we propose a conjectured cut time, and prove that it provides an upper bound on the first conjugate time, a key step in the extended Hadamard technique. In the three-dimensional case, we combine this method with a density argument to establish the conjecture.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:38:37 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Albert",
"Michael",
""
],
[
"Borza",
"Samuël",
""
],
[
"Gordina",
"Maria",
""
]
] |
2509.03412
|
David Bour
|
David A. Bour, Avishai Gilkis, Christopher A. Tout
|
Hydrogen-deficient binary stars with magnetic braking
|
11 pages, 11 figures, and 2 tables. Accepted for publication in MNRAS
| null | null | null |
astro-ph.SR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Hydrogen-deficient binary stars comprise one star which has been stripped of its hydrogen through mass transfer to a binary companion. Observations show that the companion is able to accrete several solar masses without spinning up to critical rotation, and so there must be a mechanism to drain spin angular momentum from the accretor. We test magnetically coupled winds and magnetic star-disc coupling as possible mechanisms and find that, while the disc coupling is negligible, the winds are sufficient to allow the accretor to gain mass without spinning up to critical rotation. However, in order to fully replicate observations, time-dependent scalings of the dynamo-generated magnetic field are needed.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:39:10 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Bour",
"David A.",
""
],
[
"Gilkis",
"Avishai",
""
],
[
"Tout",
"Christopher A.",
""
]
] |
2509.03413
|
Lewis Bulled
|
Lewis Bulled and Yingkai Ouyang
|
An angular momentum approach to quantum insertion errors
|
14 pages, 3 figures
| null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Quantum insertion errors are a class of errors that increase the number of qubits in a quantum system. Despite a wealth of research on classical insertion errors, there has been limited progress towards a general framework for correcting quantum insertion errors. We detail a quantum error correction protocol that can correct single insertion errors on a class of gapped permutation-invariant codes. We provide a simple two-stage syndrome extraction protocol that yields a two-bit syndrome, by measuring the total angular momentum and its projection along the $z$-axis (modulo the code gap) of the post-insertion state. We demonstrate that these measurements project the state onto a new codespace, and we detail a teleportation protocol to map the projected state back to a permutation-invariant code on the desired number of qubits.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:39:24 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Bulled",
"Lewis",
""
],
[
"Ouyang",
"Yingkai",
""
]
] |
2509.03414
|
Denis Gessert
|
Denis Gessert and Martin Weigel and Wolfhard Janke
|
The frustrated Ising model on the honeycomb lattice: Metastability and universality
|
15 pages. 12 figures, 2 tables
| null | null | null |
cond-mat.stat-mech physics.comp-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study the Ising model with competing ferromagnetic nearest- and antiferromagnetic next-nearest-neighbor interactions of strengths $J_1 > 0$ and $J_2 < 0$, respectively, on the honeycomb lattice. For $J_2 > - J_1 / 4$ it has a ferromagnetic ground state, and previous work has shown that at least for $J_2 \gtrsim -0.2 J_1$ the transition is in the Ising universality class. For even lower $J_2$ some indicators pointing towards a first-order transition were reported. By utilizing population annealing Monte Carlo simulations together with a rejection-free and adaptive update, we can equilibrate systems with $J_2$ as low as $-0.23 J_1$. By means of a finite-size scaling analysis we show that the system undergoes a second-order phase transition within the Ising universality class at least down to $J_2 =-0.23 J_1$ and, most likely, for all $J_2 > - J_1 / 4$. As we show here, there exist very long-lived metastable states in this system explaining the first-order like behavior seen in only partially equilibrated systems.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:42:34 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Gessert",
"Denis",
""
],
[
"Weigel",
"Martin",
""
],
[
"Janke",
"Wolfhard",
""
]
] |
2509.03415
|
Taekyun Kim
|
Taekyun Kim, Dae san Kim, Kyo-Shin Hwang, Dmitry V. Dolgy
|
A note on new type degenerate Srirling numbers of the first kind
|
8 pages
| null | null | null |
math.NT
|
http://creativecommons.org/licenses/by/4.0/
|
We introduce a new sequence of unsigned degenerate Stirling numbers of the first kind. Following the work of Adell-Lekuona, who represented unsigned Stirling numbers of the first kind as multiples of the expectations of specific random variables, we express our new numbers as finite sums of multiples of the expectations of certain random variables. We also provide a representation of these new numbers as finite sums involving the classical unsigned Stirling numbers of the first kind. As an inversion formula, we define a corresponding sequence of new type degenerate Stirling numbers of the second kind. We derive expressions for these numbers as finite sums that involve the Stirling numbers of the second kind.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:43:43 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Kim",
"Taekyun",
""
],
[
"Kim",
"Dae san",
""
],
[
"Hwang",
"Kyo-Shin",
""
],
[
"Dolgy",
"Dmitry V.",
""
]
] |
2509.03416
|
Moritz Kade
|
Moritz Kade
|
Integrable systems: From the ice rule to supersymmetric fishnet Feynman diagrams
|
PhD thesis, 16+131 pages, feedback and comments are very welcome
| null |
10.18452/33769
|
HU-EP-25/32-RTG
|
hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
This thesis examines the correspondence between models of statistical physics and Feynman graphs of quantum field theories (QFTs) by a common property: integrability. We review integrable structures for periodic boundary conditions on both sides, while focusing on the eight- and six-vertex model and the bi-scalar fishnet theory. The latter is a double-scaled $\gamma$-deformation of $\mathcal{N} = 4$ super Yang-Mills theory. Interesting applications of integrability existing in the literature that we reconsider are the computation of the free energy in the thermodynamic limit and its QFT counterpart, the critical coupling. In addition, we provide a detailed overview of the calculation of exact anomalous dimensions and operator product expansion (OPE) coefficients in the conformal bi-scalar fishnet theory. The original contributions of this work comprise the results of the critical coupling for models with fermions, the brick wall theory, and the fermionic fishnet theory. Additionally, we extend the study of integrable Feynman graphs to supersymmetric diagrams in superspace. By establishing an efficient graphical formalism, we obtain the critical coupling of double-scaled $\beta$-deformations of $\mathcal{N} = 4$ super Yang-Mills theory and Aharony-Bergman-Jafferis-Maldacena theory, the super brick wall and superfishnet theory, respectively. Moreover, we apply superspace methods to the superfishnet theory and find results for anomalous dimensions and an OPE coefficient, which are all-loop exact in the coupling. In addition, we study boundary integrability in the six-vertex model and for Feynman diagrams. We present new box-shaped boundary conditions for the six-vertex model and conjecture a closed form for its partition function at any lattice size. On the QFT side, we find integrable boundary scattering matrices in the form of generalized Feynman diagrams by graphical methods.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:44:16 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Kade",
"Moritz",
""
]
] |
2509.03417
|
Spyros Rigas
|
Spyros Rigas, Dhruv Verma, Georgios Alexandridis, Yixuan Wang
|
Initialization Schemes for Kolmogorov-Arnold Networks: An Empirical Study
|
30 pages, 19 figures
| null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Kolmogorov-Arnold Networks (KANs) are a recently introduced neural architecture that replace fixed nonlinearities with trainable activation functions, offering enhanced flexibility and interpretability. While KANs have been applied successfully across scientific and machine learning tasks, their initialization strategies remain largely unexplored. In this work, we study initialization schemes for spline-based KANs, proposing two theory-driven approaches inspired by LeCun and Glorot, as well as an empirical power-law family with tunable exponents. Our evaluation combines large-scale grid searches on function fitting and forward PDE benchmarks, an analysis of training dynamics through the lens of the Neural Tangent Kernel, and evaluations on a subset of the Feynman dataset. Our findings indicate that the Glorot-inspired initialization significantly outperforms the baseline in parameter-rich models, while power-law initialization achieves the strongest performance overall, both across tasks and for architectures of varying size. All code and data accompanying this manuscript are publicly available at https://github.com/srigas/KAN_Initialization_Schemes.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:45:28 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Rigas",
"Spyros",
""
],
[
"Verma",
"Dhruv",
""
],
[
"Alexandridis",
"Georgios",
""
],
[
"Wang",
"Yixuan",
""
]
] |
2509.03418
|
Jake McGrath
|
James Clarke, Jake McGrath, Colin Johnson, Jos\'e Alvarado
|
Control across scales: signals, information, and adaptive biological mechanical function
| null | null | null | null |
cond-mat.soft nlin.AO physics.bio-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Biological systems perform an astonishing array of dynamical processes -- including development and repair, regulation, behavior and motor control, sensing and signaling, and adaptation, among others. Powered by the transduction of stored energy resources, these behaviors enable biological systems to regulate functions, achieve specific outcomes, and maintain stability far from thermodynamic equilibrium. These behaviors span orders of magnitude in length and time: from nanometer-scale molecular motors driving morphogenesis to kilometer-scale seasonal migrations, and from millisecond reflexes to millennia of evolutionary adaptations. While physical laws govern the dynamics of biological systems, they alone are insufficient to fully explain how living systems sense, decide, adapt, and, ultimately, control their dynamics. In this article, we argue that control theory provides a powerful, unifying framework for understanding how biological systems regulate dynamics to maintain stability across length and time scales far from equilibrium.
|
[
{
"version": "v1",
"created": "Wed, 03 Sep 2025 15:45:35 GMT"
}
] | 2025-09-04T00:00:00 |
[
[
"Clarke",
"James",
""
],
[
"McGrath",
"Jake",
""
],
[
"Johnson",
"Colin",
""
],
[
"Alvarado",
"José",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.