id
stringlengths 9
16
| submitter
stringlengths 1
64
⌀ | authors
stringlengths 3
62.8k
| title
stringlengths 1
382
| comments
stringlengths 1
1.15k
⌀ | journal-ref
stringlengths 1
557
⌀ | doi
stringlengths 7
153
⌀ | report-no
stringlengths 1
509
⌀ | categories
stringlengths 5
125
| license
stringclasses 9
values | abstract
stringlengths 3
6.09k
| versions
listlengths 1
187
| update_date
timestamp[s]date 2007-05-23 00:00:00
2025-12-13 00:00:00
| authors_parsed
listlengths 1
3.44k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2512.10106
|
Gaurav Koley
|
Gaurav Koley and Sanika Digrajkar
|
A Simulation Framework for Studying Recommendation-Network Co-evolution in Social Platforms
| null | null | null | null |
cs.SI cs.MA
|
http://creativecommons.org/licenses/by/4.0/
|
Studying how recommendation systems reshape social networks is difficult on live platforms: confounds abound, and controlled experiments risk user harm. We present an agent-based simulator where content production, tie formation, and a graph attention network (GAT) recommender co-evolve in a closed loop. We calibrate parameters using Mastodon data and validate out-of-sample against Bluesky (4--6\% error on structural metrics; 10--15\% on held-out temporal splits). Across 18 configurations at 100 agents, we find that \emph{activation timing} affects outcomes: introducing recommendations at $t=10$ vs.\ $t=40$ decreases transitivity by 10\% while engagement differs by $<$8\%. Delaying activation increases content diversity by 9\% while reducing modularity by 4\%. Scaling experiments ($n$ up to 5,000) show the effect persists but attenuates. Jacobian analysis confirms local stability under bounded reactance parameters. We release configuration schemas and reproduction scripts.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:52:38 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Koley",
"Gaurav",
""
],
[
"Digrajkar",
"Sanika",
""
]
] |
2512.10107
|
Samuel Taylor
|
Ellis Buckminster and Samuel J. Taylor
|
Universal circles for Anosov foliations
| null | null | null | null |
math.GT
|
http://creativecommons.org/licenses/by/4.0/
|
Thurston introduced the notion of a universal circle associated to a taut foliation of a $3$-manifold as a way of organizing the ideal circle boundaries of its leaves into a single circle action. Calegari--Dunfield proved that every taut foliation of an atoroidal $3$-manifold $M$ has a universal circle, but the uniqueness (or lack-thereof) of this structure remains rather mysterious.
In this paper, we consider the foliations associated to an Anosov flow $\varphi$ on $M$, showing that several constructions of a universal circle in the literature are typically distinct. Moreover, the underlying action of the Calegari--Dunfield leftmost universal circle is generally not even conjugate to the universal circle arising from the boundary of the flow space of $\varphi$. Our primary tool is a way to use the flow space of $\varphi$ to parameterize the circle bundle at infinity of $\varphi$'s invariant foliations.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:54:08 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Buckminster",
"Ellis",
""
],
[
"Taylor",
"Samuel J.",
""
]
] |
2512.10108
|
Ali Zahra
|
Ali Zahra
|
Hydrodynamics of Multi-Species Driven Diffusive Systems with Open Boundaries: A Two-Tasep Study
| null | null | null | null |
math-ph cond-mat.stat-mech math.MP
|
http://creativecommons.org/licenses/by/4.0/
|
In this short note, we review a recently developed method for analysing multi-component driven diffusive systems with open boundaries. The approach generalises the extremal-current principle known for single-component models and is based on solving the Riemann problem for the corresponding hydrodynamic equations. As a case study, we focus on a two-species exclusion process on a lattice (Two-TASEP), where two types of particles move in opposite directions with two arbitrary rates and exchange positions upon encounter with a third rate. Despite its simplicity, this toy model effectively captures the key features of multi-species driven diffusive systems, including phase separation phenomena. This allows us to illustrate the critical role played by the underlying Riemann invariants in determining the system's macroscopic behavior.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:54:36 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zahra",
"Ali",
""
]
] |
2512.10109
|
Nizar Riane
|
Nizar Riane
|
The Moroccan Public Procurement Game
| null | null | null | null |
econ.TH math.OC
|
http://creativecommons.org/licenses/by/4.0/
|
In this paper, we study the public procurement market through the lens of game theory by modeling it as a strategic game with discontinuous and non-quasiconcave payoffs. We first show that the game admits no Nash equilibrium in pure strategies. We then analyze the two-player case and derive two explicit mixed-strategy equilibria for the symmetric game and for the weighted $(p,1-p)$ formulation. Finally, we establish the existence of a symmetric Nash equilibrium in the general $N$-player case by applying the diagonal disjoint payoff matching condition, which allows us to extend equilibrium existence to the mixed-strategy setting despite payoff discontinuities.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:57:09 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Riane",
"Nizar",
""
]
] |
2512.10110
|
Yumou Wei
|
Yumou Wei, John Stamper, Paulo F. Carvalho
|
Generate-Then-Validate: A Novel Question Generation Approach Using Small Language Models
|
Accepted as a full research paper for the 16th International Conference on Learning Analytics and Knowledge (LAK'26)
| null | null | null |
cs.CL cs.HC
|
http://creativecommons.org/licenses/by/4.0/
|
We explore the use of small language models (SLMs) for automatic question generation as a complement to the prevalent use of their large counterparts in learning analytics research. We present a novel question generation pipeline that leverages both the text generation and the probabilistic reasoning abilities of SLMs to generate high-quality questions. Adopting a "generate-then-validate" strategy, our pipeline first performs expansive generation to create an abundance of candidate questions and refine them through selective validation based on novel probabilistic reasoning. We conducted two evaluation studies, one with seven human experts and the other with a large language model (LLM), to assess the quality of the generated questions. Most judges (humans or LLMs) agreed that the generated questions had clear answers and generally aligned well with the intended learning objectives. Our findings suggest that an SLM can effectively generate high-quality questions when guided by a well-designed pipeline that leverages its strengths.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:59:36 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Wei",
"Yumou",
""
],
[
"Stamper",
"John",
""
],
[
"Carvalho",
"Paulo F.",
""
]
] |
2512.10111
|
Marianne Bauer
|
Marianne Bauer, Akshit Goyal, Sidhartha Goyal, Gautam Reddy, Shaon Chakrabarti, Michael M Desai, William Gilpin, Jacopo Grilli, Kabir Husain, Sanjay Jain, Mohit Kumar Jolly, Kyogo Kawaguchi, Aneta Koseska, Milo Lin, Leelavati Narlikar, Simone Pigolotti, Archishman Raju, Krishna Shrinivas, Rahul Siddharthan, Greg J Stephens, Andreas Tiffeau-Mayer, Suriyanarayanan Vaikuntanathan
|
Unifying Theories in High-Dimensional Biology: Approaches, Challenges and Opportunities
| null | null | null | null |
physics.bio-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Across biological subdisciplines, the last decade has seen an explosion of high-dimensional datasets, including datasets for cells, species, immune systems, neurons and behaviour. At the ICTS workshop 'Unifying Theories in High-Dimensional Biophysics' we discussed whether this high dimensionality poses a challenge or opportunity for describing, understanding and predicting biological systems theoretically. We discussed methods, models and frameworks that can help with addressing empirical observations based on these high-dimensional datasets. We summarize the challenges and opportunities that emerged in discussions according to individual participants below.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:03:31 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Bauer",
"Marianne",
""
],
[
"Goyal",
"Akshit",
""
],
[
"Goyal",
"Sidhartha",
""
],
[
"Reddy",
"Gautam",
""
],
[
"Chakrabarti",
"Shaon",
""
],
[
"Desai",
"Michael M",
""
],
[
"Gilpin",
"William",
""
],
[
"Grilli",
"Jacopo",
""
],
[
"Husain",
"Kabir",
""
],
[
"Jain",
"Sanjay",
""
],
[
"Jolly",
"Mohit Kumar",
""
],
[
"Kawaguchi",
"Kyogo",
""
],
[
"Koseska",
"Aneta",
""
],
[
"Lin",
"Milo",
""
],
[
"Narlikar",
"Leelavati",
""
],
[
"Pigolotti",
"Simone",
""
],
[
"Raju",
"Archishman",
""
],
[
"Shrinivas",
"Krishna",
""
],
[
"Siddharthan",
"Rahul",
""
],
[
"Stephens",
"Greg J",
""
],
[
"Tiffeau-Mayer",
"Andreas",
""
],
[
"Vaikuntanathan",
"Suriyanarayanan",
""
]
] |
2512.10112
|
Christian Basteck
|
Christian Basteck, Ulysse Lojkine
|
Power and Freedom in Mechanisms
| null | null | null | null |
econ.TH
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In a strategy-proof mechanism, the influence of an agent may be measured as the set of outcomes an agent can bring about by varying her (reported) type. More specifically, we refer to an agent's influence on her own relevant outcomes as her freedom, and to the influence on outcomes relevant for other agents as her power over others. The framework generalises both the notion of opportunity set from the freedom of choice literature, and established power indices for binary voting. It identifies constrained efficient mechanisms as those that maximise agents' freedom. Applying our framework to the analysis of assignment rules, we provide novel characterisations of the top trading cycles rule and bipolar serial dictatorships in terms of their freedom and power properties.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:05:01 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Basteck",
"Christian",
""
],
[
"Lojkine",
"Ulysse",
""
]
] |
2512.10113
|
Lorenzo Cima
|
Aldo Cerulli, Benedetta Tessa, Giuseppe La Selva, Oronzo Mazzeo, Lorenzo Cima, Lucia Monacis, Stefano Cresci
|
Dark Personality Traits and Online Toxicity: Linking Self-Reports to Reddit Activity
| null | null | null | null |
cs.CY cs.HC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Dark personality traits have been linked to online misbehavior such as trolling, incivility, and toxic speech. Yet the relationship between these traits and actual online conduct remains understudied. Here we investigate the associations between dark traits, online toxicity, and the socio-linguistic characteristics of online user activity. To explore this relationship, we developed a Web application that integrates validated psychological questionnaires from Amazon Mechanical Turk users to their Reddit activity data. This allowed collecting nearly 57K Reddit comments, including 2.2M tokens and 152.7K sentences from 114 users, that we systematically represent through 224 linguistic and behavioral features. We then examined their relationship to questionnaire-based trait measures via multiple correlation analyses. Among our findings is that dark traits primarily influence the production rather than the perception of online incivility. Sadistic and psychopathic tendencies are most strongly associated with overtly toxic language, whereas other dark dispositions manifest more subtly, often eluding simple textual proxies. Self-reported engagement in hostile behavior mirrors actual online activity, while existing hand-crafted textual proxies for dark triad traits show limited correspondence with our validated measures. Finally, bright and dark traits interact in nuanced ways, with extraversion reducing trolling tendencies and conscientiousness showing modest associations with entitlement and callousness. These findings deepen understanding of how personality shapes toxic online behavior and highlight both opportunities and challenges for developing reliable computational tools and targeted, effective moderation strategies.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:06:30 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Cerulli",
"Aldo",
""
],
[
"Tessa",
"Benedetta",
""
],
[
"La Selva",
"Giuseppe",
""
],
[
"Mazzeo",
"Oronzo",
""
],
[
"Cima",
"Lorenzo",
""
],
[
"Monacis",
"Lucia",
""
],
[
"Cresci",
"Stefano",
""
]
] |
2512.10114
|
Shaohu Zhang
|
Mesafint Fanuel, Mahmoud Nabil Mahmoud, Crystal Cook Marshal, Vishal Lakhotia, Biswanath Dari, Kaushik Roy, Shaohu Zhang
|
AgriRegion: Region-Aware Retrieval for High-Fidelity Agricultural Advice
|
15 pages
| null | null | null |
cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Large Language Models (LLMs) have demonstrated significant potential in democratizing access to information. However, in the domain of agriculture, general-purpose models frequently suffer from contextual hallucination, which provides non-factual advice or answers are scientifically sound in one region but disastrous in another due to variations in soil, climate, and local regulations. We introduce AgriRegion, a Retrieval-Augmented Generation (RAG) framework designed specifically for high-fidelity, region-aware agricultural advisory. Unlike standard RAG approaches that rely solely on semantic similarity, AgriRegion incorporates a geospatial metadata injection layer and a region-prioritized re-ranking mechanism. By restricting the knowledge base to verified local agricultural extension services and enforcing geo-spatial constraints during retrieval, AgriRegion ensures that the advice regarding planting schedules, pest control, and fertilization is locally accurate. We create a novel benchmark dataset, AgriRegion-Eval, which comprises 160 domain-specific questions across 12 agricultural subfields. Experiments demonstrate that AgriRegion reduces hallucinations by 10-20% compared to state-of-the-art LLMs systems and significantly improves trust scores according to a comprehensive evaluation.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:06:41 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Fanuel",
"Mesafint",
""
],
[
"Mahmoud",
"Mahmoud Nabil",
""
],
[
"Marshal",
"Crystal Cook",
""
],
[
"Lakhotia",
"Vishal",
""
],
[
"Dari",
"Biswanath",
""
],
[
"Roy",
"Kaushik",
""
],
[
"Zhang",
"Shaohu",
""
]
] |
2512.10115
|
Jorge Vila
|
Jorge Vila
|
Decoding How Proteins Fold
|
18 pages, 3 figures
| null | null | null |
q-bio.BM
|
http://creativecommons.org/licenses/by/4.0/
|
The primary aims of this perspective are twofold. Firstly, to assess whether a universal principle could govern the process of protein folding. Secondly, to clarify whether the folding of a protein could be characterized by an effective trajectory, as determined by the principle of least action, while concurrently addressing the Levinthal Paradox. The effective-trajectory conjecture, which is compatible with the existence of multiple folding pathways and characterized by a single folding rate, addresses the search timescale paradox by explaining how a protein in a proper environment always reaches its native state at a consistent folding rate and in a biologically reasonable time frame (as determined by the thermodynamic hypothesis, or Anfinsen dogma). The preliminary findings support the hypothesis that the protein folding process is governed by a universal principle: that of least action. From this viewpoint, it should be regarded as a fundamental physical model that not only facilitates the identification of folding pathways and trajectories but also offers a physical framework for understanding how nature may link the optimal kinetic speed of the protein folding process with its higher thermodynamic stability.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:06:59 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Vila",
"Jorge",
""
]
] |
2512.10116
|
Andrew Razjigaev Dr
|
Andrew Razjigaev, Hans Lohr, Alejandro Vargas-Uscategui, Peter King, Tirthankar Bandyopadhyay
|
Fast Functionally Redundant Inverse Kinematics for Robotic Toolpath Optimisation in Manufacturing Tasks
|
Published at the Australasian Conference on Robotics and Automation (ACRA 2025) https://ssl.linklings.net/conferences/acra/acra2025_proceedings/views/includes/files/pap149s2.pdf
| null | null | null |
cs.RO
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Industrial automation with six-axis robotic arms is critical for many manufacturing tasks, including welding and additive manufacturing applications; however, many of these operations are functionally redundant due to the symmetrical tool axis, which effectively makes the operation a five-axis task. Exploiting this redundancy is crucial for achieving the desired workspace and dexterity required for the feasibility and optimisation of toolpath planning. Inverse kinematics algorithms can solve this in a fast, reactive framework, but these techniques are underutilised over the more computationally expensive offline planning methods. We propose a novel algorithm to solve functionally redundant inverse kinematics for robotic manipulation utilising a task space decomposition approach, the damped least-squares method and Halley's method to achieve fast and robust solutions with reduced joint motion. We evaluate our methodology in the case of toolpath optimisation in a cold spray coating application on a non-planar surface. The functionally redundant inverse kinematics algorithm can quickly solve motion plans that minimise joint motion, expanding the feasible operating space of the complex toolpath. We validate our approach on an industrial ABB manipulator and cold-spray gun executing the computed toolpath.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:07:07 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Razjigaev",
"Andrew",
""
],
[
"Lohr",
"Hans",
""
],
[
"Vargas-Uscategui",
"Alejandro",
""
],
[
"King",
"Peter",
""
],
[
"Bandyopadhyay",
"Tirthankar",
""
]
] |
2512.10117
|
Sangli Teng
|
Sangli Teng, Hang Liu, Jingyu Song, Koushil Sreenath
|
CHyLL: Learning Continuous Neural Representations of Hybrid Systems
| null | null | null | null |
cs.LG cs.AI cs.RO cs.SY eess.SP eess.SY
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Learning the flows of hybrid systems that have both continuous and discrete time dynamics is challenging. The existing method learns the dynamics in each discrete mode, which suffers from the combination of mode switching and discontinuities in the flows. In this work, we propose CHyLL (Continuous Hybrid System Learning in Latent Space), which learns a continuous neural representation of a hybrid system without trajectory segmentation, event functions, or mode switching. The key insight of CHyLL is that the reset map glues the state space at the guard surface, reformulating the state space as a piecewise smooth quotient manifold where the flow becomes spatially continuous. Building upon these insights and the embedding theorems grounded in differential topology, CHyLL concurrently learns a singularity-free neural embedding in a higher-dimensional space and the continuous flow in it. We showcase that CHyLL can accurately predict the flow of hybrid systems with superior accuracy and identify the topological invariants of the hybrid systems. Finally, we apply CHyLL to the stochastic optimal control problem.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:07:16 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Teng",
"Sangli",
""
],
[
"Liu",
"Hang",
""
],
[
"Song",
"Jingyu",
""
],
[
"Sreenath",
"Koushil",
""
]
] |
2512.10118
|
Pol Mestres
|
Pol Mestres, Shima Sadat Mousavi, Pio Ong, Lizhi Yang, Ersin Das, Joel W. Burdick, Aaron D. Ames
|
Explicit Control Barrier Function-based Safety Filters and their Resource-Aware Computation
| null | null | null | null |
eess.SY cs.SY math.OC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper studies the efficient implementation of safety filters that are designed using control barrier functions (CBFs), which minimally modify a nominal controller to render it safe with respect to a prescribed set of states. Although CBF-based safety filters are often implemented by solving a quadratic program (QP) in real time, the use of off-the-shelf solvers for such optimization problems poses a challenge in applications where control actions need to be computed efficiently at very high frequencies. In this paper, we introduce a closed-form expression for controllers obtained through CBF-based safety filters. This expression is obtained by partitioning the state-space into different regions, with a different closed-form solution in each region. We leverage this formula to introduce a resource-aware implementation of CBF-based safety filters that detects changes in the partition region and uses the closed-form expression between changes. We showcase the applicability of our approach in examples ranging from aerospace control to safe reinforcement learning.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:09:53 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Mestres",
"Pol",
""
],
[
"Mousavi",
"Shima Sadat",
""
],
[
"Ong",
"Pio",
""
],
[
"Yang",
"Lizhi",
""
],
[
"Das",
"Ersin",
""
],
[
"Burdick",
"Joel W.",
""
],
[
"Ames",
"Aaron D.",
""
]
] |
2512.10119
|
Alessio D'Agliano
|
A. D'Agliano (1 and 2), A. V. Zlobin (3), I. Novitski (3), G. Vallone (1), P. Ferracin (1), E. Barzi (4), S. Donati (2), and V. Giusti (2) ((1) Lawrence Berkeley National Laboratory, (2) Pisa University, (3) Fermi National Accelerator Laboratory, (4) Ohio State University)
|
A Computational Procedure for Assessing I$_c$($\varepsilon$) in Nb$_3$Sn/Bi-2212 Hybrid Magnets
|
2025 International Conference on Magnet Technology (MT29)
|
IEEE Trans.Appl.Supercond. 36, 3 (2026) 8400205
|
10.1109/TASC.2025.3624725
|
FERMILAB-PUB-25-0893-V
|
physics.ins-det
|
http://creativecommons.org/licenses/by/4.0/
|
The critical current of superconductors is commonly measured by testing unloaded wires under an external magnetic field. While stressed by intense Lorentz forces, the existing HTS/LTS superconductors are prone to a reduction in critical current before reaching their structural mechanical limit. In this work, the magnetic and mechanical analysis of the FNAL 4-layer Bi-2212/Nb$_3$Sn hybrid dipole magnet is reported, aimed at predicting the critical current degradation for both the superconductors during powering at 16 T. All the Rutherford cables in the coils of the hybrid magnet were modeled at the strand level in Ansys APDL with the heterogeneous cable model. Utilizing this detailed geometry, it was possible to evaluate the effects of strain on the critical current degradation for both the Nb$_3$Sn and Bi-2212 superconductors under the intense Lorentz forces. The analysis presented in this paper integrates strain-dependent critical current laws, with parameters derived from experimental data, to simulate the hybrid magnet's performance for all possible current-powering configurations. The proposed methodology enables a detailed assessment of conductor integrity and I$_C$($\varepsilon$) reduction in existing hybrid magnet designs, providing a versatile and rigorous framework for optimizing future high-field hybrid magnets.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:12:06 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"D'Agliano",
"A.",
"",
"1 and 2"
],
[
"Zlobin",
"A. V.",
"",
"Fermi National Accelerator Laboratory"
],
[
"Novitski",
"I.",
"",
"Fermi National Accelerator Laboratory"
],
[
"Vallone",
"G.",
"",
"Lawrence Berkeley National Laboratory"
],
[
"Ferracin",
"P.",
"",
"Lawrence Berkeley National Laboratory"
],
[
"Barzi",
"E.",
"",
"Ohio State University"
],
[
"Donati",
"S.",
"",
"Pisa University"
],
[
"Giusti",
"V.",
"",
"Pisa University"
]
] |
2512.10120
|
Maris Basha
|
Maris Basha, Anja Zai, Sabine Stoll, Richard Hahnloser
|
VocSim: A Training-free Benchmark for Zero-shot Content Identity in Single-source Audio
| null | null | null | null |
cs.SD cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
General-purpose audio representations aim to map acoustically variable instances of the same event to nearby points, resolving content identity in a zero-shot setting. Unlike supervised classification benchmarks that measure adaptability via parameter updates, we introduce VocSim, a training-free benchmark probing the intrinsic geometric alignment of frozen embeddings. VocSim aggregates 125k single-source clips from 19 corpora spanning human speech, animal vocalizations, and environmental sounds. By restricting to single-source audio, we isolate content representation from the confound of source separation. We evaluate embeddings using Precision@k for local purity and the Global Separation Rate (GSR) for point-wise class separation. To calibrate GSR, we report lift over an empirical permutation baseline. Across diverse foundation models, a simple pipeline, frozen Whisper encoder features, time-frequency pooling, and label-free PCA, yields strong zero-shot performance. However, VocSim also uncovers a consistent generalization gap. On blind, low-resource speech, local retrieval drops sharply. While performance remains statistically distinguishable from chance, the absolute geometric structure collapses, indicating a failure to generalize to unseen phonotactics. As external validation, our top embeddings predict avian perceptual similarity, improve bioacoustic classification, and achieve state-of-the-art results on the HEAR benchmark. We posit that the intrinsic geometric quality measured here proxies utility in unlisted downstream applications. We release data, code, and a public leaderboard to standardize the evaluation of intrinsic audio geometry.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:13:12 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Basha",
"Maris",
""
],
[
"Zai",
"Anja",
""
],
[
"Stoll",
"Sabine",
""
],
[
"Hahnloser",
"Richard",
""
]
] |
2512.10121
|
Zhongjie Jiang Sr
|
Zhongjie Jiang
|
Workflow is All You Need: Escaping the "Statistical Smoothing Trap" via High-Entropy Information Foraging and Adversarial Pacing
|
22 pages, 8 figures. Includes an ecological validity blind test where the Agentic Workflow achieved a 25% acceptance rate in top-tier media, decisively outperforming the SOTA Zero-shot baseline (0%). Features the DNFO-v5 ontology
| null | null | null |
cs.CL cs.AI cs.CY q-fin.GN
|
http://creativecommons.org/licenses/by/4.0/
|
Central to long-form text generation in vertical domains is the "impossible trinity" confronting current large language models (LLMs): the simultaneous achievement of low hallucination, deep logical coherence, and personalized expression. This study establishes that this bottleneck arises from existing generative paradigms succumbing to the Statistical Smoothing Trap, a phenomenon that overlooks the high-entropy information acquisition and structured cognitive processes integral to expert-level writing. To address this limitation, we propose the DeepNews Framework, an agentic workflow that explicitly models the implicit cognitive processes of seasoned financial journalists. The framework integrates three core modules: first, a dual-granularity retrieval mechanism grounded in information foraging theory, which enforces a 10:1 saturated information input ratio to mitigate hallucinatory outputs; second, schema-guided strategic planning, a process leveraging domain expert knowledge bases (narrative schemas) and Atomic Blocks to forge a robust logical skeleton; third, adversarial constraint prompting, a technique deploying tactics including Rhythm Break and Logic Fog to disrupt the probabilistic smoothness inherent in model-generated text. Experiments delineate a salient Knowledge Cliff in deep financial reporting: content truthfulness collapses when retrieved context falls below 15,000 characters, while a high-redundancy input exceeding 30,000 characters stabilizes the Hallucination-Free Rate (HFR) above 85%. In an ecological validity blind test conducted with a top-tier Chinese technology media outlet, the DeepNews system--built on a previous-generation model (DeepSeek-V3-0324)-achieved a 25% submission acceptance rate, significantly outperforming the 0% acceptance rate of zero-shot generation by a state-of-the-art (SOTA) model (GPT-5).
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:13:55 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Jiang",
"Zhongjie",
""
]
] |
2512.10122
|
Hannah Potgieter
|
Hannah Potgieter, Razvan C. Fetecau, Steven J. Ruuth
|
Numerical approximation of the first $p$-Laplace eigenpair
| null | null | null | null |
math.NA cs.NA math.SP
|
http://creativecommons.org/licenses/by/4.0/
|
We approximate the first Dirichlet eigenpair of the $p$-Laplace operator for $2 \leq p < \infty$ on both Euclidean and surface domains. We emphasize large $p$ values and discuss how the $p \to \infty$ limit connects to the underlying geometry of our domain. Working with large $p$ values introduces significant numerical challenges. We present a surface finite element numerical scheme that combines a Newton inverse-power iteration with a new domain rescaling strategy, which enables stable computations for large $p$. Numerical experiments in $1$D, planar domains, and surfaces embedded in $\mathbb{R}^3$ demonstrate the accuracy and robustness of our approach and show convergence towards the $p \to \infty$ limiting behavior.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:14:47 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Potgieter",
"Hannah",
""
],
[
"Fetecau",
"Razvan C.",
""
],
[
"Ruuth",
"Steven J.",
""
]
] |
2512.10123
|
Olivia Tsang
|
Olivia Tsang, Owen Melia, Vasileios Charisopoulos, Jeremy Hoskins, Yuehaw Khoo, Rebecca Willett
|
A Model-Guided Neural Network Method for the Inverse Scattering Problem
|
28 pages
| null | null | null |
physics.comp-ph cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Inverse medium scattering is an ill-posed, nonlinear wave-based imaging problem arising in medical imaging, remote sensing, and non-destructive testing. Machine learning (ML) methods offer increased inference speed and flexibility in capturing prior knowledge of imaging targets relative to classical optimization-based approaches; however, they perform poorly in regimes where the scattering behavior is highly nonlinear. A key limitation is that ML methods struggle to incorporate the physics governing the scattering process, which are typically inferred implicitly from the training data or loosely enforced via architectural design. In this paper, we present a method that endows a machine learning framework with explicit knowledge of problem physics, in the form of a differentiable solver representing the forward model. The proposed method progressively refines reconstructions of the scattering potential using measurements at increasing wave frequencies, following a classical strategy to stabilize recovery. Empirically, we find that our method provides high-quality reconstructions at a fraction of the computational or sampling costs of competing approaches.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:15:58 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Tsang",
"Olivia",
""
],
[
"Melia",
"Owen",
""
],
[
"Charisopoulos",
"Vasileios",
""
],
[
"Hoskins",
"Jeremy",
""
],
[
"Khoo",
"Yuehaw",
""
],
[
"Willett",
"Rebecca",
""
]
] |
2512.10124
|
Erqian Cai
|
Dorian W. P. Amaral, Erqian Cai, Andrew J. Long, Juehang Qin, and Christopher D. Tunnell
|
Towards the Direct Detection of Composite Ultraheavy Dark Matter in Quantum Sensor Arrays
|
25+5 pages, 8 figures, 1 table, 2 appendices
| null | null | null |
hep-ph hep-ex
|
http://creativecommons.org/licenses/by/4.0/
|
Quantum sensor arrays have recently been proposed as a promising platform for the direct detection of ultraheavy dark matter, which is typically assumed to behave as a point-like particle. However, particles with masses at or above the Planck scale cannot be elementary; instead, they must exist as composite objects with finite spatial extent. Such spatially extended dark matter models lead to distinctive phenomenology in these detectors, particularly when the dark matter also interacts through long-range forces with their own characteristic length scales. In this work, we study the sensitivity of quantum sensor arrays to composite, ultraheavy dark matter interacting via both gravity and a novel Yukawa force. We consider three phenomenologically motivated density profiles -- a tophat, a Gaussian, and an exponential -- and contrast their signals with the point-like limit. Using a Monte Carlo analysis based on the predicted impulse signals and estimates of thermal and quantum noise, we obtain sensitivity projections for a future realization of a quantum sensor array. We find a non-trivial interplay between the dark-matter scale radius, the inter-sensor spacing, and the Yukawa screening length. Future accelerometer arrays would provide valuable information about the mass and size of composite ultraheavy dark matter, and our work will help to characterize the signatures of different theoretical models of ultraheavy dark matter.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:18:21 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Amaral",
"Dorian W. P.",
""
],
[
"Cai",
"Erqian",
""
],
[
"Long",
"Andrew J.",
""
],
[
"Qin",
"Juehang",
""
],
[
"Tunnell",
"Christopher D.",
""
]
] |
2512.10125
|
Kyungtae Park
|
Avidit Acharya, Kyungtae Park, Tomer Zaidman
|
Motivated Reasoning and Information Aggregation
| null | null | null | null |
econ.TH
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
If agents engage in motivated reasoning, how does that affect the aggregation of information in society? We study the effects of motivated reasoning in two canonical settings - the Condorcet jury theorem (CJT), and the sequential social learning model (SLM). We define a notion of motivated reasoning that applies to these and a broader class of other settings, and contrast it to other approaches in the literature. We show for the CJT that information aggregates in the large electorate limit even with motivated reasoning. When signal quality differs across states, increasing motivation improves welfare in the state with the more informative signal and worsens it in the other state. In the SLM, motivated reasoning improves information aggregation up to a point; but if agents place too little weight on truth-seeking, this can lead to worse aggregation relative to the fully Bayesian benchmark.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:20:36 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Acharya",
"Avidit",
""
],
[
"Park",
"Kyungtae",
""
],
[
"Zaidman",
"Tomer",
""
]
] |
2512.10126
|
Bret Underwood
|
S. Shajidul Haque, Ghadir Jafari, Bret Underwood
|
Inflation is Not Magic
|
17 pages, 1 figure
| null | null | null |
hep-th gr-qc quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Cosmological perturbations generated during inflation exhibit striking quantum features, including entanglement and high circuit complexity. Yet their observational signatures remain effectively indistinguishable from classical stochastic variables. We quantify this tension by showing that quantum inflationary perturbations are continuous variable stabilizer states with vanishing quantum magic, a necessary resource for universal quantum computation as measured by Wigner negativity. Consequently, despite their quantum origins and description, these states can be efficiently simulated using classical algorithms. We further show that the Wigner negativity arising from primordial non-Gaussianity is suppressed not only by the non-linearity parameter $f_{NL}$, but also by the exponential squeezing of the perturbations. Viewing the early universe as a "high complexity, low magic" regime provides another perspective of what it means for the origin of structure in the universe to be "quantum."
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:21:30 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Haque",
"S. Shajidul",
""
],
[
"Jafari",
"Ghadir",
""
],
[
"Underwood",
"Bret",
""
]
] |
2512.10127
|
Tanmoy Bhattacharya
|
Tanmoy Bhattacharya (Los Alamos National Laboratory)
|
Comment on the claim of physical irrelevance of the topological term
|
Comment on arXiv:2001.07152
| null | null |
LA-UR-22-29996
|
hep-lat
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We argue that the claim of the absence of charge-conjugation-parity (CP) violating effects due to a topological term in quantum chromodynamics (QCD) is based upon a misunderstanding, and the standard results in the field are correct.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:21:55 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Bhattacharya",
"Tanmoy",
"",
"Los Alamos National Laboratory"
]
] |
2512.10128
|
Chuan Huang
|
Chuan Huang, Gustaf Hendeby, Isaac Skog
|
Inertial Magnetic SLAM Systems Using Low-Cost Sensors
| null | null | null | null |
cs.RO eess.SP
|
http://creativecommons.org/licenses/by/4.0/
|
Spatially inhomogeneous magnetic fields offer a valuable, non-visual information source for positioning. Among systems leveraging this, magnetic field-based simultaneous localization and mapping (SLAM) systems are particularly attractive because they can provide positioning information and build a magnetic field map on the fly. Moreover, they have bounded error within mapped regions. However, state-of-the-art methods typically require low-drift odometry data provided by visual odometry or a wheel encoder, etc. This is because these systems need to minimize/reduce positioning errors while exploring, which happens when they are in unmapped regions. To address these limitations, this work proposes a loosely coupled and a tightly coupled inertial magnetic SLAM (IM-SLAM) system. The proposed systems use commonly available low-cost sensors: an inertial measurement unit (IMU), a magnetometer array, and a barometer. The use of non-visual data provides a significant advantage over visual-based systems, making it robust to low-visibility conditions. Both systems employ state-space representations, and magnetic field models on different scales. The difference lies in how they use a local and global magnetic field model. The loosely coupled system uses these models separately in two state-space models, while the tightly coupled system integrates them into one state-space model. Experiment results show that the tightly coupled IM-SLAM system achieves lower positioning errors than the loosely coupled system in most scenarios, with typical errors on the order of meters per 100 meters traveled. These results demonstrate the feasiblity of developing a full 3D IM-SLAM systems using low-cost sensors and the potential of applying these systems in emergency response scenarios such as mine/fire rescue.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:22:00 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Huang",
"Chuan",
""
],
[
"Hendeby",
"Gustaf",
""
],
[
"Skog",
"Isaac",
""
]
] |
2512.10129
|
David Gao
|
David Gao and David Jekel
|
Elementary embeddings into ultrapower $\mathrm{II}_1$ factors without a UCP lift
|
8 pages. Comments welcome
| null | null | null |
math.OA math.LO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We show that there are $\mathrm{II}_1$ factors $M$ and elementary embeddings $M \to M^{\mathcal{U}}$ which do not lift to sequences of UCP maps, and in fact $M$ can be chosen from any given elementary equivalence class. Furthermore, under continuum hypothesis, we show that in the sense of cardinality "most" automorphisms of a ultrapower $M^{\mathcal{U}}$ of a separable $\mathrm{II}_1$ factor do not lift to a sequence of UCP maps $\varphi_n: M \to M$.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:22:38 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Gao",
"David",
""
],
[
"Jekel",
"David",
""
]
] |
2512.10130
|
Jack Foley
|
Jack Foley, Alice Shapley, Ryan Sanders, Naveen A. Reddy, Michael W. Topping, Thomas M. Stanton, Max Pettini, Fergus Cullen, Richard S. Ellis, N. M. F\"orster Schreiber, Tucker Jones, Anthony J. Pahl, Leonardo Clarke, Natalie Lam
|
The AURORA Survey: Constraining Chemical Enrichment Pathways at Cosmic Noon with Argon Abundances
|
16 pages, 4 figures, 1 table, submitted to ApJ
| null | null | null |
astro-ph.GA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present argon abundances from a sample of 46 star-forming galaxies at $z=2-3.5$ from the Assembly of Ultradeep Rest-Optical Observations Revealing Astrophysics (AURORA) program. Although argon is an $\alpha-$element produced by Core Collapse Supernovae (CCSNe), the latest supernova yield models suggest additional argon production and enrichment by Type Ia supernovae (SNe Ia), unlike other $\alpha-$elements such as oxygen. To study the relationship between argon and oxygen abundances, we construct two median-stacked composite spectra for separate $z=2.0-2.6$ and $z=2.8-3.5$ redshift bins, presenting more representative measurements than previous samples that require individual detection of faint lines. Abundance ratios were determined using an empirical calibration based on the strength of the [ArIII]$\lambda7137$ emission line relative to the [OIII]$\lambda5008$ emission line. With this calibration, we estimate argon abundances (Ar/O) of $0.42^{+0.12}_{-0.10}\mbox{(Ar/O)}_{\odot}$ for the $\langle z \rangle = 2.26$ bin and $0.42^{+0.12}_{-0.11}\mbox{(Ar/O)}_{\odot}$ for the $\langle z \rangle = 3.15$ bin, suggesting minimal SNe Ia and dominant CCSNe enrichment in this sample. Comparison of our abundance measurements of $z\sim 2-3$ AURORA galaxies with chemical evolution modeling of Milky Way stars shows consistency with the Milky Way Bulge component, suggesting a rapid star-formation timescale. However, even larger samples of actively star-forming galaxies with available argon abundances, as well as comparisons between argon abundance and other critical galaxy properties (e.g., sSFR) and models (e.g., one tuned specifically to this redshift range) are needed to draw stronger conclusions on the role of argon in galactic chemical enrichment at Cosmic Noon.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:22:52 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Foley",
"Jack",
""
],
[
"Shapley",
"Alice",
""
],
[
"Sanders",
"Ryan",
""
],
[
"Reddy",
"Naveen A.",
""
],
[
"Topping",
"Michael W.",
""
],
[
"Stanton",
"Thomas M.",
""
],
[
"Pettini",
"Max",
""
],
[
"Cullen",
"Fergus",
""
],
[
"Ellis",
"Richard S.",
""
],
[
"Schreiber",
"N. M. Förster",
""
],
[
"Jones",
"Tucker",
""
],
[
"Pahl",
"Anthony J.",
""
],
[
"Clarke",
"Leonardo",
""
],
[
"Lam",
"Natalie",
""
]
] |
2512.10131
|
Pietro Chiarantoni
|
Pietro Chiarantoni, Oscar Serra, Mohammad Erfan Mowlaei, Venkata Surya Kumar Choutipalli, Mark DelloStritto, Xinghua Shi, Micheal L. Klein, Vincenzo Carnevale
|
Generative Modeling of Entangled Polymers with a Distance-Based Variational Autoencoder
| null | null | null | null |
physics.comp-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present a variational autoencoder framework for learning and generating configurations of structured polymer globules from distance matrices. We used coarse-grained molecular dynamics to sample polyethylene structures, which we used as the training set for our deep learning model. By combining convolution and attention layers, the model encodes the structural patterns of distance matrices into an organized and roto-translationally invariant latent space of lower dimensionality. The generative capability of the variational autoencoder, coupled with a post-processing pipeline based on multidimensional scaling and short molecular dynamics, enables the recovery of physically meaningful configurations. The reconstructed and generated samples reproduce key observables, including energy, size, and entanglement, despite minor discrepancies in the raw decoder output.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:25:07 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chiarantoni",
"Pietro",
""
],
[
"Serra",
"Oscar",
""
],
[
"Mowlaei",
"Mohammad Erfan",
""
],
[
"Choutipalli",
"Venkata Surya Kumar",
""
],
[
"DelloStritto",
"Mark",
""
],
[
"Shi",
"Xinghua",
""
],
[
"Klein",
"Micheal L.",
""
],
[
"Carnevale",
"Vincenzo",
""
]
] |
2512.10132
|
Logan Nye
|
Logan Nye
|
Universal Hirschberg for Width Bounded Dynamic Programs
|
31 pages
| null | null | null |
cs.DS cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Hirschberg's algorithm (1975) reduces the space complexity for the longest common subsequence problem from $O(N^2)$ to $O(N)$ via recursive midpoint bisection on a grid dynamic program (DP). We show that the underlying idea generalizes to a broad class of dynamic programs with local dependencies on directed acyclic graphs (DP DAGs). Modeling a DP as deterministic time evolution over a topologically ordered DAG with frontier width $\omega$ and bounded in-degree, and assuming a max-type semiring with deterministic tie breaking, we prove that in a standard offline random-access model any such DP admits deterministic traceback in space $O(\omega \log T + (\log T)^{O(1)})$ cells over a fixed finite alphabet, where $T$ is the number of states. Our construction replaces backward dynamic programs by forward-only recomputation and organizes the time order into a height-compressed recursion tree whose nodes expose small "middle frontiers'' across which every optimal path must pass. The framework yields near-optimal traceback bounds for asymmetric and banded sequence alignment, one-dimensional recurrences, and dynamic-programming formulations on graphs of bounded pathwidth. We also show that an $\Omega(\omega)$ space term (in bits) is unavoidable in forward single-pass models and discuss conjectured $\sqrt{T}$-type barriers in streaming settings, supporting the view that space-efficient traceback is a structural property of width-bounded DP DAGs rather than a peculiarity of grid-based algorithms.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:26:22 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Nye",
"Logan",
""
]
] |
2512.10133
|
Gabriel Francisco Alves Bastos
|
Gabriel F.A. Bastos and Jugurta Montalv\~ao
|
Partitioning the Sample Space for a More Precise Shannon Entropy Estimation
|
The manuscript contains 6 pages and 10 figures. It has been accepted for International Conference on Artificial Intelligence, Computer, Data Sciences and Applications (ACDSA 2026)
| null | null | null |
cs.LG math.ST stat.TH
|
http://creativecommons.org/licenses/by/4.0/
|
Reliable data-driven estimation of Shannon entropy from small data sets, where the number of examples is potentially smaller than the number of possible outcomes, is a critical matter in several applications. In this paper, we introduce a discrete entropy estimator, where we use the decomposability property in combination with estimations of the missing mass and the number of unseen outcomes to compensate for the negative bias induced by them. Experimental results show that the proposed method outperforms some classical estimators in undersampled regimes, and performs comparably with some well-established state-of-the-art estimators.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:26:36 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Bastos",
"Gabriel F. A.",
""
],
[
"Montalvão",
"Jugurta",
""
]
] |
2512.10134
|
Ryan Mann
|
Ryan L. Mann, Gabriel Waite
|
Approximate Counting in Local Lemma Regimes
|
20 pages, 0 figures
| null | null | null |
cs.DS math.CO math.PR quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We establish efficient approximate counting algorithms for several natural problems in local lemma regimes. In particular, we consider the probability of intersection of events and the dimension of intersection of subspaces. Our approach is based on the cluster expansion method. We obtain fully polynomial-time approximation schemes for both the probability of intersection and the dimension of intersection for commuting projectors. For general projectors, we provide two algorithms: a fully polynomial-time approximation scheme under a global inclusion-exclusion stability condition, and an efficient affine approximation under a spectral gap assumption. As corollaries of our results, we obtain efficient algorithms for approximating the number of satisfying assignments of conjunctive normal form formulae and the dimension of satisfying subspaces of quantum satisfiability formulae.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:26:36 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Mann",
"Ryan L.",
""
],
[
"Waite",
"Gabriel",
""
]
] |
2512.10135
|
Hubert Djuitcheu
|
Hubert Djuitcheu, Andrew Sergeev, Khurshid Alam, Danny Santhosh, Achim Autenrieth, Jochen Seitz
|
Lightweight Security for Private Networks: Real-World Evaluation of WireGuard
| null | null | null | null |
cs.CR cs.NI hep-ex
|
http://creativecommons.org/licenses/by/4.0/
|
This paper explores WireGuard as a lightweight alternative to IPsec for securing the user plane as well as the control plane in an industrial Open RAN deployment at the Adtran Terafactory in Meiningen. We focus on a realistic scenario where external vendors access their hardware in our 5G factory network, posing recurrent security risks from untrusted gNBs and intermediate network elements. Unlike prior studies limited to lab setups, we implement a complete proof-of-concept in a factory environment and compare WireGuard with IPsec under industrial traffic conditions. Our approach successfully protects user data (N3 interface) against untrusted gNBs and man-in-the-middle attacks while enabling control plane (N2 interface) authentication between the access and mobility management functions (AMF) and gNB. Performance measurements show that WireGuard adds minimal overhead in throughput, latency, and Central Processing Unit (CPU) usage, achieving performance comparable to IPsec. These findings demonstrate that WireGuard offers competitive performance with significantly reduced configuration complexity, making it a strong candidate for broader adoption in O-RAN, providing a unified, lightweight security layer across multiple interfaces and components.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:30:00 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Djuitcheu",
"Hubert",
""
],
[
"Sergeev",
"Andrew",
""
],
[
"Alam",
"Khurshid",
""
],
[
"Santhosh",
"Danny",
""
],
[
"Autenrieth",
"Achim",
""
],
[
"Seitz",
"Jochen",
""
]
] |
2512.10136
|
Sebastian Munoz
|
Max Engelstein, Inwon Kim, Sebastian Munoz
|
Regularity of the free boundary for the supercooled Stefan problem in arbitrary dimensions
|
44 pages
| null | null | null |
math.AP
|
http://creativecommons.org/licenses/by/4.0/
|
We study the free boundary in the supercooled Stefan problem, a classical model for the solidification of water below its freezing temperature. In contrast with the melting problem, physical experiments and heuristics indicate that the water--ice interface in the supercooled problem may exhibit fractal freezing sets, infinite-speed propagation of the frozen front, and nucleation (the spontaneous appearance of ice). Despite this, we show that the free boundary has a robust structure.
We decompose the free boundary into three parts: (1) a regular part that advances with finite speed in time; (2) a singular part consisting of points where the front attains infinite speed or nucleates, but with controlled space-time (i.e., $\leq d-1$ parabolic) dimension; and (3) a jump component, which can have large dimension in a time slice, but which is contained in a space-time smooth graph and occurs only at a zero-dimensional set of times. Examples show that each of these parts can be nonempty.
Furthermore, we prove that the free boundary is the graph $t=s(x)$ of a continuously differentiable freezing time $s$, and the singular set coincides with the critical set of $s$, proving that singularities in supercooled freezing always occur with infinite speed.
These results provide the first free boundary regularity theory for the supercooled Stefan problem in arbitrary dimensions.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:32:13 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Engelstein",
"Max",
""
],
[
"Kim",
"Inwon",
""
],
[
"Munoz",
"Sebastian",
""
]
] |
2512.10137
|
Kondo Gnanvo
|
Kondo Gnanvo (for the PICOSEC Collaboration)
|
{\mu}RWELL-PICOSEC: Precision Timing with Resistive Micro-Well Detector
| null | null | null | null |
physics.ins-det
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
The PICOSEC detector concept uses a micro-pattern gaseous detector (MPGD) amplification structure combined with a Cerenkov radiator coated with a semi-transparent photocathode to provide below tens of picosecond-level precision timing capabilities with minimum ionizing particles. PICOSEC has triggered interest in the development of time-of-flight detectors for particle identification and timing detectors for track reconstruction in the high rate environment of future nuclear and high energy physics experiments. The PICOSEC Micromegas (or PICOSEC-MM) detector, developed by the CERN-based PICOSEC collaboration, use the Micromegas structure for gaseous amplification and achieve below 20 ps timing resolution. A new type of PICOSEC detector, the {\mu}RWELL-PICOSEC based on {\mu}RWELL amplification structure, is being investigated at Thomas Jefferson National Accelerator Facility (Jefferson Lab) alongside PICOSEC-MM R&D efforts in Europe. Preliminary results from the two 2024 beam test campaigns at CERN demonstrate a timing performance better than 24 ps is achievable with a single-channel {\mu}RWELL-PICOSEC prototype. A vigorous R&D effort is ongoing to improve the timing performance, robustness and operational stability of {\mu}RWELL-PICOSEC detectors. Development of a large size {\mu}RWELL-PICOSEC is also under consideration for applications in large scale experiments.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:33:38 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Gnanvo",
"Kondo",
"",
"for the PICOSEC Collaboration"
]
] |
2512.10138
|
Raymond Chu
|
Raymond Chu, Inwon Kim, Sebastian Munoz
|
The supercooled Stefan problem: fractal freezing and the fine structure of maximal solutions
| null | null | null | null |
math.AP math.PR
|
http://creativecommons.org/licenses/by/4.0/
|
We study the supercooled Stefan problem in arbitrary dimensions. First, we study general solutions and their irregularities, showing generic fractal freezing and nucleation, based on a novel Markovian gluing principle. In contrast, we then establish regularity properties of maximal solutions, which are obtained by maximizing a suitable notion of "average" freezing time. Unexpectedly, we show that maximal solutions have a transition zone that is open modulo a low-dimensional set: this allows us to apply obstacle problem theory for a finer regularity analysis. We further show that maximal solutions are in general non-universal, and we obtain sharp stability results under perturbation of each maximal solution. Lastly, we study maximal solutions in both the radial and the one-dimensional setting. We show that in these cases the maximal solution is universal and minimizes nucleation, in agreement with phenomena observed in the physics literature.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:37:46 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chu",
"Raymond",
""
],
[
"Kim",
"Inwon",
""
],
[
"Munoz",
"Sebastian",
""
]
] |
2512.10139
|
Jin Sun
|
Jin Sun and Kui Wang
|
Parabolic Frequency on Gaussian Spaces and Unique Continuation
|
15 pages. Comments are welcome
| null | null | null |
math.AP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We establish an almost-monotonicity formula for a parabolic frequency on Gaussian spaces for solutions of the Ornstein-Uhlenbeck heat equation with lower-order terms: $$\partial_t u = L_\gamma u + b(x,t) \cdot \nabla u + c(x,t)u, $$ where $L_\gamma = \Delta - x \cdot \nabla$ is the Ornstein-Uhlenbeck operator. In contrast to classical results that require $b$ and $c$ to be bounded, we only assume that $b$ is bounded and $c$ satisfies a linear growth condition, while the solution $u$ is allowed to have at most exponential quadratic growth. The key innovation is a weighted $L^2$ framework that uses the backward Mehler kernel as a weight, which naturally encodes the underlying measure and compensates for the unbounded coefficients. From the frequency monotonicity, we derive the strong unique continuation principle. This extends Poon's seminal results and complements recent geometric generalizations by Colding and Minicozzi in the context of Gaussian measure spaces. We further apply our framework to establish unique continuation for equations with potentials exhibiting quadratic growth or certain singularities.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:38:57 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Sun",
"Jin",
""
],
[
"Wang",
"Kui",
""
]
] |
2512.10140
|
Ryan Camacho
|
Siavash Mirzaei-Ghormish, Mahmoud Jalali Mehrabad, Helaman Flores, Dirk Englund, Ryan M. Camacho
|
Programmable Quantum Photonic Interfaces for Quantum Networking
| null | null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Quantum networks require interfaces translating memory photons to telecom wavelengths while controlling spatial modes; tasks performed by separate components today. We present a programmable alternative: a structured pump writes a virtual Bragg grating enabling simultaneous spatio-spectral conversion and real-time controlling of emission. Using a LiNbO$_3$ whispering-gallery resonator, we demonstrate 93\% spatial coupling and bidirectional conversion between 736\,nm and 1347\,nm. This reconfigurable interface eliminates cascaded losses and hardware modifications.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:44:48 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Mirzaei-Ghormish",
"Siavash",
""
],
[
"Mehrabad",
"Mahmoud Jalali",
""
],
[
"Flores",
"Helaman",
""
],
[
"Englund",
"Dirk",
""
],
[
"Camacho",
"Ryan M.",
""
]
] |
2512.10141
|
Sarwan Ali
|
Sarwan Ali, Taslim Murad, Imdadullah Khan
|
Sequence-to-Image Transformation for Sequence Classification Using Rips Complex Construction and Chaos Game Representation
| null | null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Traditional feature engineering approaches for molecular sequence classification suffer from sparsity issues and computational complexity, while deep learning models often underperform on tabular biological data. This paper introduces a novel topological approach that transforms molecular sequences into images by combining Chaos Game Representation (CGR) with Rips complex construction from algebraic topology. Our method maps sequence elements to 2D coordinates via CGR, computes pairwise distances, and constructs Rips complexes to capture both local structural and global topological features. We provide formal guarantees on representation uniqueness, topological stability, and information preservation. Extensive experiments on anticancer peptide datasets demonstrate superior performance over vector-based, sequence language models, and existing image-based methods, achieving 86.8\% and 94.5\% accuracy on breast and lung cancer datasets, respectively. The topological representation preserves critical sequence information while enabling effective utilization of vision-based deep learning architectures for molecular sequence analysis.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:46:15 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ali",
"Sarwan",
""
],
[
"Murad",
"Taslim",
""
],
[
"Khan",
"Imdadullah",
""
]
] |
2512.10142
|
S. C. Park
|
S. C. Park
|
Conformal Invariance of the FK-Ising Model on Lorentz-Maximal S-Embeddings
|
14 pages, 1 figure
| null | null | null |
math.PR math-ph math.MP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We show on non-flat but critical s-embeddings the celebrated convergence of the interface curves of the critical FK Ising model to an $\operatorname{SLE}_{16/3}$ curve, using discrete complex analytic techniques first used in arXiv:0708.0039, arXiv:1312.0533 and subsequently extended to more lattice settings including isoradial graphs arXiv:0910.2045, circle packings arXiv:1712.08736, and flat s-embeddings arXiv:2006.14559. In our setting, the s-embedding approximates a maximal surface in the Minkowski space $\mathbb R^{2,1}$, an `exact' criticality condition identified in arXiv:2006.14559, which is stronger than the percolation-theoretic `near-critical' setup studied in, e.g., arXiv:2309.08470. The proof relies on a careful discretisation of the Laplace-Beltrami operator on the s-embedding, which is crucial in identifying the limit of the martingale observable.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:49:40 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Park",
"S. C.",
""
]
] |
2512.10143
|
Igor Petrashkevich
|
Igor Petrashkevich, Anna Punanova, Paola Caselli, Jaime E. Pineda, Olli Sipila, Anton I. Vasyunin
|
Deuterium fractionation and CO depletion in Barnard 5
|
accepted for publication in ApJ
| null | null | null |
astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
Deuterium fractionation provides a key diagnostic of the physical and chemical evolution of prestellar and protostellar cores, where it is strongly linked to CO depletion in cold, dense gas. We present the first spatially resolved maps of deuterium fraction and CO depletion in the Barnard 5 (B5) region of the Perseus molecular cloud, covering both a starless core and the protostellar core hosting the Class 0/I source IRAS 03445+3242. Using IRAM 30~m observations of N$_2$H$^+$(1--0), N$_2$D$^+$(1--0), H$^{13}$CO$^+$(1--0), and DCO$^+$(2--1), complemented by C$^{18}$O(2--1) data, we derive column density, deuterium fraction, and CO depletion maps. We find that the deuterium fraction in both mentioned nitrogen- and carbon-bearing species increases from the protostellar to the starless core, reaching $R_D^{\rm N_2H^+}=0.43\pm0.10$ and $R_D^{\rm HCO^+}=0.09\pm0.02$ in the starless core, compared with $0.15\pm0.03$ and $0.05\pm0.01$, respectively, in the protostellar core. The CO depletion factor also rises from $4.1\pm0.1$ to $5.0\pm0.1$ across the same transition. While the embedded YSO reduces deuteration in the dense inner gas, the less dense envelope traced by HCO$^+$ is only slightly affected at our resolution. Our analysis confirms that CO freeze-out and the presence of a protostar jointly regulate deuterium chemistry in star-forming regions.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:52:19 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Petrashkevich",
"Igor",
""
],
[
"Punanova",
"Anna",
""
],
[
"Caselli",
"Paola",
""
],
[
"Pineda",
"Jaime E.",
""
],
[
"Sipila",
"Olli",
""
],
[
"Vasyunin",
"Anton I.",
""
]
] |
2512.10144
|
Chun Hei Leung
|
Chun Hei Leung and Pak-Tik Fong and Tianyi Yan and Weibin Li
|
Engineer coherent oscillatory modes in Markovian open quantum systems
|
12 pages, 4 figures
| null | null | null |
quant-ph cond-mat.quant-gas
|
http://creativecommons.org/licenses/by/4.0/
|
We develop a novel framework to engineer persistent oscillatory modes in Markovian open quantum systems governed by a time-independent Lindblad master equation. We show that oscillatory modes can be created when the Hamiltonian and jump operator can be expressed in the same block-diagonal form. A key feature of the framework is that the dissipator of the Lindblad master equation are generally non-zero. We identify the weak and strong conditions, where the onset of the oscillatory modes is dependent and independent of the parameters of the system, respectively. Our method extends beyond the typical decoherence-free subspace approach, in which the dissipator is zero. We demonstrate the applicability of this framework using various models, showing how carefully tailored system-environment interactions can produce sustained coherent oscillations.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:52:59 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Leung",
"Chun Hei",
""
],
[
"Fong",
"Pak-Tik",
""
],
[
"Yan",
"Tianyi",
""
],
[
"Li",
"Weibin",
""
]
] |
2512.10145
|
Qing-Dong Jiang
|
Qing-Dong Jiang
|
Harnessing Vacuum Fluctuations to Shape Electronic and Photonic Behavior
|
Perspective, 5 pages
|
npj Nanophotonics 2, 46 (2025)
|
10.1038/s44310-025-00091-4
| null |
cond-mat.mes-hall physics.optics
|
http://creativecommons.org/licenses/by/4.0/
|
Vacuum quantum fluctuations are an inescapable and fundamental feature of modern physics. By integrating cavity-enhanced or surface-modified vacuum quantum fluctuations with low-dimensional materials, a new paradigm-vacuumronics-emerges, enabling unprecedented control over both material properties and photonic responses at the micro- and nanoscale. This synergy opens novel pathways for engineering quantum light-matter interactions, advancing applications in quantum photonics, nanoscale optoelectronics, and quantum material design.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:53:09 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Jiang",
"Qing-Dong",
""
]
] |
2512.10146
|
Eliezer Estrecho
|
Yow-Ming (Robin) Hu, Mateusz Kr\'ol, Daria A. Smirnova, Lev A. Smirnov, Bianca Rae Fabricante, Karol Winkler, Martin Kamp, Christian Schneider, Sven H\"ofling, Timothy C. H. Liew, Andrew G. Truscott, Elena A. Ostrovskaya, and Eliezer Estrecho
|
Momentum-space non-Hermitian skin effect in an exciton-polariton system
|
19 pages, 17 figures
| null | null | null |
physics.optics cond-mat.mes-hall
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Localization of a macroscopic number of eigenstates on a real-space boundary, known as the non-Hermitian skin effect, is one of the striking topological features emerging from non-Hermiticity. Realizing this effect typically requires periodic (lattice) systems with asymmetry of intersite coupling, which is not readily available in many physical platforms. Instead, it is meticulously engineered, e.g., in photonics, which results in complex structures requiring precise fabrication steps. Here, we propose a simpler mechanism: introducing an asymmetric, purely imaginary potential in a topologically trivial system induces momentum-space localization akin to the skin effect. We experimentally demonstrate this localization using exciton polaritons, hybrid light-matter quasi-particles in a simple engineered `round box' trap, pumped by a laser pump offset from the trap center. The effect disappears if the pump is concentric with the trap. The localization persists and becomes stronger at higher densities of polaritons, when a non-equilibrium Bose-Einstein condensate is formed and the system becomes nonlinear. Our approach offers a new route to realizing skin effects in continuous, non-periodic systems and exploring the interplay of non-Hermiticity, topology, and nonlinearity in macroscopic quantum states.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 22:56:53 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Yow-Ming",
"",
"",
"Robin"
],
[
"Hu",
"",
""
],
[
"Król",
"Mateusz",
""
],
[
"Smirnova",
"Daria A.",
""
],
[
"Smirnov",
"Lev A.",
""
],
[
"Fabricante",
"Bianca Rae",
""
],
[
"Winkler",
"Karol",
""
],
[
"Kamp",
"Martin",
""
],
[
"Schneider",
"Christian",
""
],
[
"Höfling",
"Sven",
""
],
[
"Liew",
"Timothy C. H.",
""
],
[
"Truscott",
"Andrew G.",
""
],
[
"Ostrovskaya",
"Elena A.",
""
],
[
"Estrecho",
"Eliezer",
""
]
] |
2512.10147
|
Sarwan Ali
|
Sarwan Ali, Taslim Murad
|
Murmur2Vec: A Hashing Based Solution For Embedding Generation Of COVID-19 Spike Sequences
| null | null | null | null |
cs.LG q-bio.GN
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Early detection and characterization of coronavirus disease (COVID-19), caused by SARS-CoV-2, remain critical for effective clinical response and public-health planning. The global availability of large-scale viral sequence data presents significant opportunities for computational analysis; however, existing approaches face notable limitations. Phylogenetic tree-based methods are computationally intensive and do not scale efficiently to today's multi-million-sequence datasets. Similarly, current embedding-based techniques often rely on aligned sequences or exhibit suboptimal predictive performance and high runtime costs, creating barriers to practical large-scale analysis. In this study, we focus on the most prevalent SARS-CoV-2 lineages associated with the spike protein region and introduce a scalable embedding method that leverages hashing to generate compact, low-dimensional representations of spike sequences. These embeddings are subsequently used to train a variety of machine learning models for supervised lineage classification. We conduct an extensive evaluation comparing our approach with multiple baseline and state-of-the-art biological sequence embedding methods across diverse metrics. Our results demonstrate that the proposed embeddings offer substantial improvements in efficiency, achieving up to 86.4\% classification accuracy while reducing embedding generation time by as much as 99.81\%. This highlights the method's potential as a fast, effective, and scalable solution for large-scale viral sequence analysis.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:03:10 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ali",
"Sarwan",
""
],
[
"Murad",
"Taslim",
""
]
] |
2512.10148
|
Moonsoo Park
|
Moonsoo Park, Jeongseok Yun, Bohyung Kim
|
PARAN: Persona-Augmented Review ANswering system on Food Delivery Review Dataset
| null | null | null | null |
cs.CL cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Personalized review response generation presents a significant challenge in domains where user information is limited, such as food delivery platforms. While large language models (LLMs) offer powerful text generation capabilities, they often produce generic responses when lacking contextual user data, reducing engagement and effectiveness. In this work, we propose a two-stage prompting framework that infers both explicit (e.g., user-stated preferences) and implicit (e.g., demographic or stylistic cues) personas directly from short review texts. These inferred persona attributes are then incorporated into the response generation prompt to produce user-tailored replies. To encourage diverse yet faithful generations, we adjust decoding temperature during inference. We evaluate our method using a real-world dataset collected from a Korean food delivery app, and assess its impact on precision, diversity, and semantic consistency. Our findings highlight the effectiveness of persona-augmented prompting in enhancing the relevance and personalization of automated responses without requiring model fine-tuning.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:04:48 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Park",
"Moonsoo",
""
],
[
"Yun",
"Jeongseok",
""
],
[
"Kim",
"Bohyung",
""
]
] |
2512.10149
|
Han Chen
|
Han Chen, Steven Zhu, Yingrui Li
|
STARS: Semantic Tokens with Augmented Representations for Recommendation at Scale
| null | null | null | null |
cs.IR cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Real-world ecommerce recommender systems must deliver relevant items under strict tens-of-milliseconds latency constraints despite challenges such as cold-start products, rapidly shifting user intent, and dynamic context including seasonality, holidays, and promotions. We introduce STARS, a transformer-based sequential recommendation framework built for large-scale, low-latency ecommerce settings. STARS combines several innovations: dual-memory user embeddings that separate long-term preferences from short-term session intent; semantic item tokens that fuse pretrained text embeddings, learnable deltas, and LLM-derived attribute tags, strengthening content-based matching, long-tail coverage, and cold-start performance; context-aware scoring with learned calendar and event offsets; and a latency-conscious two-stage retrieval pipeline that performs offline embedding generation and online maximum inner-product search with filtering, enabling tens-of-milliseconds response times. In offline evaluations on production-scale data, STARS improves Hit@5 by more than 75 percent relative to our existing LambdaMART system. A large-scale A/B test on 6 million visits shows statistically significant lifts, including Total Orders +0.8%, Add-to-Cart on Home +2.0%, and Visits per User +0.5%. These results demonstrate that combining semantic enrichment, multi-intent modeling, and deployment-oriented design can yield state-of-the-art recommendation quality in real-world environments without sacrificing serving efficiency.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:16:02 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chen",
"Han",
""
],
[
"Zhu",
"Steven",
""
],
[
"Li",
"Yingrui",
""
]
] |
2512.10150
|
Hani Itani
|
Lama Alssum, Hani Itani, Hasan Abed Al Kader Hammoud, Philip Torr, Adel Bibi, Bernard Ghanem
|
Unforgotten Safety: Preserving Safety Alignment of Large Language Models with Continual Learning
| null | null | null | null |
cs.CL cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The safety alignment of large language models (LLMs) is becoming increasingly important with their democratization. In this paper, we study the safety degradation that comes with adapting LLMs to new tasks. We attribute this safety compromise to catastrophic forgetting and frame the problem of preserving safety when fine-tuning as a continual learning (CL) problem. We consider the fine-tuning-as-a-service setup where the user uploads their data to a service provider to get a customized model that excels on the user's selected task. We adapt several CL approaches from the literature and systematically evaluate their ability to mitigate safety degradation. These include regularization-based, memory-based, and model merging approaches. We consider two scenarios, (1) benign user data and (2) poisoned user data. Our results demonstrate that CL approaches consistently achieve lower attack success rates than standard fine-tuning. Among these, DER outperforms both other CL methods and existing safety-preserving baselines while maintaining task utility. These findings generalize across three downstream tasks (GSM8K, SST2, Code) and three model families (LLaMA2-7B, Mistral-7B, Gemma-2B), establishing CL as a practical solution to preserve safety.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:16:47 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Alssum",
"Lama",
""
],
[
"Itani",
"Hani",
""
],
[
"Hammoud",
"Hasan Abed Al Kader",
""
],
[
"Torr",
"Philip",
""
],
[
"Bibi",
"Adel",
""
],
[
"Ghanem",
"Bernard",
""
]
] |
2512.10151
|
Charles Fanning
|
Charles Fanning, Mehmet Emin Aktas
|
Topological Conditioning for Mammography Models via a Stable Wavelet-Persistence Vectorization
|
8 Pages, 2 Figures, submitted to IEEE Transactions on Medical Imaging
| null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Breast cancer is the most commonly diagnosed cancer in women and a leading cause of cancer death worldwide. Screening mammography reduces mortality, yet interpretation still suffers from substantial false negatives and false positives, and model accuracy often degrades when deployed across scanners, modalities, and patient populations. We propose a simple conditioning signal aimed at improving external performance based on a wavelet based vectorization of persistent homology. Using topological data analysis, we summarize image structure that persists across intensity thresholds and convert this information into spatial, multi scale maps that are provably stable to small intensity perturbations. These maps are integrated into a two stage detection pipeline through input level channel concatenation. The model is trained and validated on the CBIS DDSM digitized film mammography cohort from the United States and evaluated on two independent full field digital mammography cohorts from Portugal (INbreast) and China (CMMD), with performance reported at the patient level. On INbreast, augmenting ConvNeXt Tiny with wavelet persistence channels increases patient level AUC from 0.55 to 0.75 under a limited training budget.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:19:05 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Fanning",
"Charles",
""
],
[
"Aktas",
"Mehmet Emin",
""
]
] |
2512.10152
|
Tiago Brogueira Mr
|
Tiago Brogueira, M\'ario Figueiredo
|
Rethinking Causal Discovery Through the Lens of Exchangeability
|
37 pages, 4 figures
| null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Causal discovery methods have traditionally been developed under two distinct regimes: independent and identically distributed (i.i.d.) and timeseries data, each governed by separate modelling assumptions. In this paper, we argue that the i.i.d. setting can and should be reframed in terms of exchangeability, a strictly more general symmetry principle. We present the implications of this reframing, alongside two core arguments: (1) a conceptual argument, based on extending the dependency of experimental causal inference on exchangeability to causal discovery; and (2) an empirical argument, showing that many existing i.i.d. causal-discovery methods are predicated on exchangeability assumptions, and that the sole extensive widely-used real-world "i.i.d." benchmark (the T\"ubingen dataset) consists mainly of exchangeable (and not i.i.d.) examples. Building on this insight, we introduce a novel synthetic dataset that enforces only the exchangeability assumption, without imposing the stronger i.i.d. assumption. We show that our exchangeable synthetic dataset mirrors the statistical structure of the real-world "i.i.d." dataset more closely than all other i.i.d. synthetic datasets. Furthermore, we demonstrate the predictive capability of this dataset by proposing a neural-network-based causal-discovery algorithm trained exclusively on our synthetic dataset, and which performs similarly to other state-of-the-art i.i.d. methods on the real-world benchmark.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:19:39 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Brogueira",
"Tiago",
""
],
[
"Figueiredo",
"Mário",
""
]
] |
2512.10153
|
Newshaw Bahreyni
|
Newshaw Bahreyni, Paul M. Alsing, Carlo Cafaro, Walid Redjem, Christian Corda
|
Upper Bounds on Fluctuation Growths of Observables in Open Quantum Systems
|
18 pages, 1 figure
| null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The upper bounds for the rate of fluctuation growth of an observable in both open and closed quantum systems have been studied actively recently. In our recent work we showed that the rate of fluctuation growth for an observable in a closed quantum system is upper bounded by the fluctuation of its corresponding velocity-like observable. That bound also indicated a tradeoff between the time derivatives of the mean and the standard deviation. In this paper we will look at open quantum systems in two cases. For the first case we find the generator of evolution for an open system employing both the Taylor expansion and the standard time-ordered evolution via the Dyson series, while in the second case we consider no specific information about the evolution of the system. We then find the rate of fluctuation growth in each case. Comparing the upper bounds for each case and considering the upper bound found for a closed system suggest that including more details by separating the contributions of the system and state dynamics seems to result in looser bounds for the rate of fluctuation growth.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:21:13 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Bahreyni",
"Newshaw",
""
],
[
"Alsing",
"Paul M.",
""
],
[
"Cafaro",
"Carlo",
""
],
[
"Redjem",
"Walid",
""
],
[
"Corda",
"Christian",
""
]
] |
2512.10154
|
Masato Fujita
|
Masato Fujita
|
Cardinality of the sets of dimension functions in ordered structures
| null | null | null | null |
math.LO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We compute the cardinality $\mathfrak n_{\dim}(\mathcal M)$ of the sets of dimension functions on the ordered structures $\mathcal M$. The inequality $\mathfrak n_{\dim}(\mathcal M) \leq 1$ holds if $\mathcal M$ is a d-minimal expansion of an ordered group. If $\mathcal M$ is o-minimal and $\mathfrak n_{\dim}(\mathcal M)<\infty$, there exists a positive integer $m$ such that $\mathfrak n_{\dim}(\mathcal M)=2^m-1$. For every positive integer $m$, there exists a weakly o-minimal expansion $\mathcal M$ of an ordered divisible Abelian group such that $\mathfrak n_{\dim}(\mathcal M)=m$.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:24:36 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Fujita",
"Masato",
""
]
] |
2512.10155
|
Jeongeun Kim
|
Jeongeun Kim and Christopher Torng
|
A Vertically Integrated Framework for Templatized Chip Design
| null | null | null | null |
cs.AR cs.SE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Developers who primarily engage with software often struggle to incorporate custom hardware into their applications, even though specialized silicon can provide substantial benefits to machine learning and AI, as well as to the application domains that they enable. This work investigates how a chip can be generated from a high-level object-oriented software specification, targeting introductory-level chip design learners with only very light performance requirements, while maintaining mental continuity between the chip layout and the software source program. In our approach, each software object is represented as a corresponding region on the die, producing a one-to-one structural mapping that preserves these familiar abstractions throughout the design flow. To support this mapping, we employ a modular construction strategy in which vertically composed IP blocks implement the behavioral protocols expressed in software. A direct syntactic translation, however, cannot meet hardware-level efficiency or communication constraints. For this reason, we leverage formal type systems based on sequences that check whether interactions between hardware modules adhere to the communication patterns described in the software model. We further examine hardware interconnect strategies for composing many such modules and develop layout techniques suited to this object-aligned design style. Together, these contributions preserve mental continuity from software to chip design for new learners and enables practical layout generation, ultimately reducing the expertise required for software developers to participate in chip creation.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:32:50 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Kim",
"Jeongeun",
""
],
[
"Torng",
"Christopher",
""
]
] |
2512.10156
|
Davud Rostam-Afschar
|
Jan Kemper and Davud Rostam-Afschar
|
Inference for Batched Adaptive Experiments
| null | null | null | null |
econ.EM cs.LG stat.ME
|
http://creativecommons.org/licenses/by/4.0/
|
The advantages of adaptive experiments have led to their rapid adoption in economics, other fields, as well as among practitioners. However, adaptive experiments pose challenges for causal inference. This note suggests a BOLS (batched ordinary least squares) test statistic for inference of treatment effects in adaptive experiments. The statistic provides a precision-equalizing aggregation of per-period treatment-control differences under heteroskedasticity. The combined test statistic is a normalized average of heteroskedastic per-period z-statistics and can be used to construct asymptotically valid confidence intervals. We provide simulation results comparing rejection rates in the typical case with few treatment periods and few (or many) observations per batch.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:33:08 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Kemper",
"Jan",
""
],
[
"Rostam-Afschar",
"Davud",
""
]
] |
2512.10157
|
Deyu Lu
|
Xuance Jiang, Jennifer Cano, Yuan Ping, Yafis Barlas and Deyu Lu
|
Two-dimensional helical superconductivity and gapless superconducting edge modes in the 1T$^\prime$-WS$_2$/2H-WS$_2$ heterophase bilayer
| null | null | null | null |
cond-mat.supr-con cond-mat.mtrl-sci
|
http://creativecommons.org/licenses/by/4.0/
|
We propose a material platform comprised of transition metal dichalcogenide (TMDC) heterostructures to realize the two-dimensional (2D) helical superconductivity with an intrinsic gap. By van der Waals stacking a 2D superconductor (1T$^\prime$-WS$_2$ with inversion symmetry) on top of a 2D topological insulator (2H-WS$_2$ with mirror symmetry), the resulting TMDC bilayer exhibits Rashba superconductivity. Under an external in-plane magnetic field, the system can host finite-momentum Cooper pairing, evidenced by the divergence in the particle-particle susceptibility of a $k\cdot p$ Hamiltonian fitted to the \textit{ab initio} theory band structure. The resulting 2D helical superconducting phase can induce superconductivity in the edge states with its spatially varying order parameter. By varying the strength of the in-plane magnetic field, we demonstrate that the helical edge state can undergo a phase transition to a one-dimensional gapless phase with narrow Fermi segments corresponding to zero-energy Bogoliubov quasi-particles. The controllable one-dimensional gapless phase serves as a clear experimental fingerprint of 2D helical superconductivity. The proposed 2D TMDC heterostructure is promising for intrinsic nonreciprocal superconducting transport and the development of Majorana-based quantum devices.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:37:01 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Jiang",
"Xuance",
""
],
[
"Cano",
"Jennifer",
""
],
[
"Ping",
"Yuan",
""
],
[
"Barlas",
"Yafis",
""
],
[
"Lu",
"Deyu",
""
]
] |
2512.10158
|
Masato Fujita
|
Masato Fujita
|
Dimension functions in weakly o-minimal structures admitting strong cell decomposition
| null | null | null | null |
math.LO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We investigate the cardinality $\mathfrak n_{\dim}(\mathcal M)$ of the sets of dimension functions on weakly o-minimal structures $\mathcal M$ admitting strong cell decomposition.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:37:39 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Fujita",
"Masato",
""
]
] |
2512.10159
|
Liangliang Chen
|
Liangliang Chen, Weiyu Sun, Ying Zhang
|
Enhancing Large Language Models for End-to-End Circuit Analysis Problem Solving
| null | null | null | null |
cs.CY cs.AI cs.HC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Large language models (LLMs) have shown strong performance in data-rich domains such as programming, but their reliability in engineering tasks remains limited. Circuit analysis -- requiring multimodal understanding and precise mathematical reasoning -- highlights these challenges. Although Gemini 2.5 Pro improves diagram interpretation and analog-circuit reasoning, it still struggles to consistently produce correct solutions when given both text and circuit diagrams. At the same time, engineering education needs scalable AI tools capable of generating accurate solutions for tasks such as automated homework feedback and question-answering. This paper presents an enhanced, end-to-end circuit problem solver built on Gemini 2.5 Pro. We first benchmark Gemini on a representative set of undergraduate circuit problems and identify two major failure modes: 1) circuit-recognition hallucinations, particularly incorrect source polarity detection, and 2) reasoning-process hallucinations, such as incorrect current directions. To address recognition errors, we integrate a fine-tuned YOLO detector and OpenCV processing to isolate voltage and current sources, enabling Gemini to re-identify source polarities from cropped images with near-perfect accuracy. To reduce reasoning errors, we introduce an ngspice-based verification loop in which Gemini generates a .cir file, ngspice simulates the circuit, and discrepancies trigger iterative regeneration with optional human-in-the-loop review. Across 83 problems, the proposed pipeline achieves a 97.59% success rate (81 correct solutions), substantially outperforming Gemini 2.5 Pro's original 79.52% accuracy. This system extends LLM capabilities for multimodal engineering problem-solving and supports the creation of high-quality educational datasets and AI-powered instructional tools.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:38:14 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chen",
"Liangliang",
""
],
[
"Sun",
"Weiyu",
""
],
[
"Zhang",
"Ying",
""
]
] |
2512.10160
|
Gavril Farkas
|
Marian Aprodu, Gavril Farkas, Claudiu Raicu, Alexander I. Suciu
|
The effective Chen ranks conjecture
|
46 pages
| null | null | null |
math.AG math.GR math.GT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Koszul modules and their associated resonance schemes are objects appearing in a variety of contexts in algebraic geometry, topology, and combinatorics. We present a proof of an effective version of the Chen ranks conjecture describing the Hilbert function of any Koszul module verifying natural conditions inspired by geometry. We give applications to hyperplane arrangements, describing in a uniform effective manner the Chen ranks of the fundamental group of the complement of every arrangement whose projective resonance is reduced. Finally, we formulate a sharp generic vanishing conjecture for Koszul modules and present a parallel between this statement and the Prym--Green Conjecture on syzygies of general Prym canonical curves.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:39:29 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Aprodu",
"Marian",
""
],
[
"Farkas",
"Gavril",
""
],
[
"Raicu",
"Claudiu",
""
],
[
"Suciu",
"Alexander I.",
""
]
] |
2512.10161
|
Jes\'us Alejandro Miranda Hern\'andez
|
Emilio J. Estrada, Alejandro Miranda, Pablo Roig
|
Hadronic contributions to $a_\mu$ within Resonance Chiral Theory
|
32 pages, 1 figure, 1 table. Invited review. Version accepted for publication in EPJ ST
| null | null | null |
hep-ph hep-ex hep-lat
|
http://creativecommons.org/licenses/by/4.0/
|
We review the recent progress achieved, using Resonance Chiral Theory, in the hadronic contributions to the muon anomalous magnetic moment. These include the hadronic vacuum polarization, either using $e^+e^-$ or $\tau$ decays into hadron final states as input; and the hadronic light-by-light part, where in addition to previous results on the lightest pseudoscalar and tensor-poles contributions, we first present the evaluation of the pseudoscalar box using this formalism. We also discuss the scalar, axial-pole and other subleading pieces. The results obtained are consistent with the White Paper 2 values, with comparable precision.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:40:25 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Estrada",
"Emilio J.",
""
],
[
"Miranda",
"Alejandro",
""
],
[
"Roig",
"Pablo",
""
]
] |
2512.10162
|
Denis Seletskiy
|
Yang Ding, Shenghao Zhang, Alexander R. Albrecht, Zhaojie Feng, Lars Forberger, Hiroki Tanaka, Markus P. Hehlen, Galina Nemova, Peter J. Pauzauskie, Denis V. Seletskiy, and Masaru Kuno
|
Solid-state Laser Cooling
|
30 pages, 4 figures, 6 tables
| null | null | null |
physics.optics physics.app-ph physics.ins-det
|
http://creativecommons.org/licenses/by/4.0/
|
Since the first proof-of-concept demonstrations of photoluminescence-based optical refrigeration, solid-state laser cooling has developed into a credible competitor to conventional cryogenic technologies. Solid-state laser cooling continues to advance as new materials push cooling limits. These developments have created a need to consolidate progress made to date as well as standardize critical experimental considerations needed for reliable and verifiable cooling measurements. This primer therefore outlines essential concepts and requirements, which underpin solid-state laser cooling. The primer summarizes key milestones achieved with cooling-grade, rare-earth-doped glasses and crystals as well as with semiconductors. It additionally highlights emerging applications of solid-state optical refrigeration. To strengthen the consistency and reproducibility of cooling results going forward, two reporting checklists are introduced. They cover materials, cooling metrics, and thermometry. This primer is intended to serve as both a tutorial and a practical reference for incoming and existing researchers involved in solid-state laser-cooling.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:41:01 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ding",
"Yang",
""
],
[
"Zhang",
"Shenghao",
""
],
[
"Albrecht",
"Alexander R.",
""
],
[
"Feng",
"Zhaojie",
""
],
[
"Forberger",
"Lars",
""
],
[
"Tanaka",
"Hiroki",
""
],
[
"Hehlen",
"Markus P.",
""
],
[
"Nemova",
"Galina",
""
],
[
"Pauzauskie",
"Peter J.",
""
],
[
"Seletskiy",
"Denis V.",
""
],
[
"Kuno",
"Masaru",
""
]
] |
2512.10163
|
Sapna Mishra
|
Sapna Mishra, Vikram Khaire, Romeo Pallikkara, Anand Narayanan, and Andrew J. Fox
|
Discovery of Weak O VI Absorption in Underdense Regions of the Low-Redshift Intergalactic Medium
|
Submitted to ApJ. Comments are welcomed
| null | null | null |
astro-ph.GA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We search for weak O VI absorption in the low-redshift intergalactic medium (IGM) using 82 high signal-to-noise quasar spectra obtained with the Cosmic Origins Spectrograph on board the Hubble Space Telescope. From this dataset, we compile a clean sample of 396 intervening Lyman-alpha (Lya) absorption lines with H I column densities log (N_HI) < 14.5, all of which lack individual O VI absorption with log (N_OVI ) > 13. We perform a spectral stacking analysis at the expected location of the O VI doublet, revealing O VI absorption with a statistical significance greater than 5$\sigma$, and measure an equivalent width of 1.7 $\pm$ 0.3 mA, corresponding to log (N_OVI ) = 12.14 $\pm$ 0.08. The stacked O VI absorption signal associated with strong Lya absorbers (13.5 <= log N_HI < 14.5) is significantly stronger than that associated with weaker Lya absorbers (12.5 <= log N_HI < 13.5). For the subset of 81 broad Lya absorbers (BLAs; b(HI) > 45 km/s), we obtain a marginal $\sim$3 $\sigma$ O VI detection. Other than Si III, detected at 5$\sigma$, no associated metal lines are found. Cross-correlation of the Lya absorbers with galaxies indicates that 93% of these absorbers are not associated with bright galaxies within 1 Mpc, implying that the detected O VI originates in the diffuse IGM rather than the circumgalactic medium. The stacked O VI signal suggests characteristic metallicities of $\sim 0.01\,Z_{\odot}$ under photoionisation and $\sim 0.001\,Z_{\odot}$ under collisional ionisation conditions, though these estimates are model-dependent and assume that O VI and H I trace the same phase. This study provides the first observational evidence for metal absorption in low-column-density Lya systems that individually exhibit no detectable metals, placing important constraints on the metal enrichment of the underdense IGM.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:48:27 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Mishra",
"Sapna",
""
],
[
"Khaire",
"Vikram",
""
],
[
"Pallikkara",
"Romeo",
""
],
[
"Narayanan",
"Anand",
""
],
[
"Fox",
"Andrew J.",
""
]
] |
2512.10164
|
Ma\"elle Magnan
|
Maelle Magnan, Tobias Geron, Izzy L. Garland, Chris J. Lintott, Jason Shingirai Makechemu, David O Ryan, Brooke D. Simmons, Rebecca J. Smethurst
|
Strong Bars, Strong Inflow: The Effect of Bar Strength on Gas Inflow
| null | null | null | null |
astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
Stellar bars are elongated structures in disk galaxies that can torque and funnel gas inward, influencing galaxy evolution. While strong bars are known to induce rapid inflow, the impact of weaker bars remains less certain. We collected spectroscopic data using the Isaac Newton Telescope to analyze 18 nearby galaxies (strongly barred, weakly barred, and unbarred) drawn from Galaxy Zoo DESI. We obtained spatial profiles of equivalent width (EW) and ionized gas velocity dispersion by fitting Gaussian profiles to the H{\alpha} emission line. Strongly barred galaxies exhibit a distinctive three-peaked EW[H{\alpha}] structure, consistent with inward funneling of gas. Weakly barred systems lack this pattern, which suggests limited inflow. Velocity dispersion distributions further distinguish the bar types, with strongly barred galaxies showing significantly higher values than weakly barred and unbarred systems. These results suggest that strong bars drive gas inflow, while weak bars exert a limited dynamical influence.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:49:59 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Magnan",
"Maelle",
""
],
[
"Geron",
"Tobias",
""
],
[
"Garland",
"Izzy L.",
""
],
[
"Lintott",
"Chris J.",
""
],
[
"Makechemu",
"Jason Shingirai",
""
],
[
"Ryan",
"David O",
""
],
[
"Simmons",
"Brooke D.",
""
],
[
"Smethurst",
"Rebecca J.",
""
]
] |
2512.10165
|
Melanie Walsh
|
Matt Miller, Dan Sinykin, Melanie Walsh
|
BookReconciler: An Open-Source Tool for Metadata Enrichment and Work-Level Clustering
|
Published in the proceedings of the Joint Conference on Digital Libraries (JCDL) 2025, Resources
|
Joint Conference on Digital Libraries (JCDL), 2025
| null | null |
cs.DL cs.IR
|
http://creativecommons.org/licenses/by/4.0/
|
We present BookReconciler, an open-source tool for enhancing and clustering book data. BookReconciler allows users to take spreadsheets with minimal metadata, such as book title and author, and automatically 1) add authoritative, persistent identifiers like ISBNs 2) and cluster related Expressions and Manifestations of the same Work, e.g., different translations or editions. This enhancement makes it easier to combine related collections and analyze books at scale. The tool is currently designed as an extension for OpenRefine -- a popular software application -- and connects to major bibliographic services including the Library of Congress, VIAF, OCLC, HathiTrust, Google Books, and Wikidata. Our approach prioritizes human judgment. Through an interactive interface, users can manually evaluate matches and define the contours of a Work (e.g., to include translations or not). We evaluate reconciliation performance on datasets of U.S. prize-winning books and contemporary world fiction. BookReconciler achieves near-perfect accuracy for U.S. works but lower performance for global texts, reflecting structural weaknesses in bibliographic infrastructures for non-English and global literature. Overall, BookReconciler supports the reuse of bibliographic data across domains and applications, contributing to ongoing work in digital libraries and digital humanities.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:51:55 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Miller",
"Matt",
""
],
[
"Sinykin",
"Dan",
""
],
[
"Walsh",
"Melanie",
""
]
] |
2512.10166
|
Khushiyant Khushiyant
|
Khushiyant
|
Emergent Collective Memory in Decentralized Multi-Agent AI Systems
|
23 pages, 4 figures
| null | null | null |
cs.MA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We demonstrate how collective memory emerges in decentralized multi-agent systems through the interplay between individual agent memory and environmental trace communication. Our agents maintain internal memory states while depositing persistent environmental traces, creating a spatially distributed collective memory without centralized control. Comprehensive validation across five environmental conditions (20x20 to 50x50 grids, 5-20 agents, 50 runs per configuration) reveals a critical asymmetry: individual memory alone provides 68.7% performance improvement over no-memory baselines (1563.87 vs 927.23, p < 0.001), while environmental traces without memory fail completely. This demonstrates that memory functions independently but traces require cognitive infrastructure for interpretation. Systematic density-sweep experiments (rho in [0.049, 0.300], up to 625 agents) validate our theoretical phase transition prediction. On realistic large grids (30x30, 50x50), stigmergic coordination dominates above rho ~ 0.20, with traces outperforming memory by 36-41% on composite metrics despite lower food efficiency. The experimental crossover confirms the predicted critical density rho_c = 0.230 within 13% error.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:54:22 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Khushiyant",
"",
""
]
] |
2512.10167
|
Anwesh Majumder
|
Anwesh Majumder, T. Heckman, J. Meunier, A. Simionescu, B.R. McNamara, L. Gu, A. Ptak, E. Hodges-Kluck, M. Yukita, M.W. Wise, N. Roy
|
Spectrally Resolved Gas Kinematics in Cygnus A: XRISM Detects AGN Jet-induced Velocity Dispersion in Multi-temperature Gas
|
Accepted for publication in ApJ
| null | null | null |
astro-ph.HE astro-ph.CO astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
We report spectral analysis on a 170 ks XRISM \textit{Resolve} exposure of the core of Cygnus A. Analyzing the full field of view spectrum in the $1.7-12.0$ keV band, we find evidence for two-temperature cluster gas. The hotter ($kT = 5.53 \pm 0.13$ keV) gas has a velocity dispersion of $261 \pm 13$ km s$^{-1}$ and a bulk velocity of $120 \pm 20$ km s$^{-1}$ with respect to the central galaxy. The cooler gas ($kT = 2.0^{+0.4}_{-0.3}$ keV) has an even broader velocity dispersion of $440 \pm 130$ km s$^{-1}$, with a systematic uncertainty of $120$ km s$^{-1}$. The relative line-of-sight velocity between the hotter and cooler gas can be as high as $450 \pm 140$ km s$^{-1}$. We interpret the high velocity dispersions as a combination of turbulence and bulk motion due to the cocoon shock. The upper limit on the non-thermal pressure fraction for the hotter gas is $7.7 \pm 0.7\%$. We associate the cooler gas with the central region ($<35$ kpc) and the hotter phase with the gas surrounding it ($35-100$ kpc). The total energy due to the kinetic motion is $5.1 \times 10^{60}$ erg, consistent with the energy associated with the central radio source. The kinetic energy injection rate is $6.9 \times 10^{44}-7.4 \times 10^{45}$ erg s$^{-1}$ under varying assumptions of injection timescales. The range of injection power is higher than the cooling luminosity, and thus the heating and cooling rates in Cygnus A are unbalanced.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:57:04 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Majumder",
"Anwesh",
""
],
[
"Heckman",
"T.",
""
],
[
"Meunier",
"J.",
""
],
[
"Simionescu",
"A.",
""
],
[
"McNamara",
"B. R.",
""
],
[
"Gu",
"L.",
""
],
[
"Ptak",
"A.",
""
],
[
"Hodges-Kluck",
"E.",
""
],
[
"Yukita",
"M.",
""
],
[
"Wise",
"M. W.",
""
],
[
"Roy",
"N.",
""
]
] |
2512.10168
|
Satoru Ohgata
|
Satoru Ohgata and Satoru Hayami
|
Large Anomalous Hall Effect in Topologically Trivial Double-$Q$ Magnets
|
Main text: 5 pages, 3 figures. Supplementary material: 7 pages, 7 figures
| null | null | null |
cond-mat.str-el
|
http://creativecommons.org/licenses/by/4.0/
|
Multi-$Q$ magnets consist of superposed spin density waves with distinct magnetic modulation vectors, enabling a wide range of magnetic orders depending on their combination. Among them, topologically nontrivial spin textures, such as a magnetic skyrmion, has been extensively studied owing to the emergence of topological Hall effects induced by real-space scalar spin chirality. Contrary to this expectation, we theoretically investigate another route to enhancing the Hall response under a topologically \textit{trivial} double-$Q$ spin textures. Despite the cancellation of the scalar spin chirality, the double-$Q$ magnetism exhibits a pronounced Hall response with a nonmonotonic dependence on the uniform magnetization, which is in stark contrast to a ferromagnetic state and a single-$Q$ spiral state. Analyzing the multi-orbital Kondo lattice model, we show that orbital hybridization induced by the double-$Q$ superstructure enhances the Berry curvature in $\mathbf{k}$-space, leading to a large anomalous Hall effect. This mechanism accounts for the observed giant anomalous Hall effect in GdRu$_2$Si$_2$ and GdRu$_2$Ge$_2$, thereby highlighting topologically trivial double-$Q$ spin textures as promising spintronic materials.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 23:58:38 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ohgata",
"Satoru",
""
],
[
"Hayami",
"Satoru",
""
]
] |
2512.10169
|
Rishi Bommasani
|
Alexander Wan, Kevin Klyman, Sayash Kapoor, Nestor Maslej, Shayne Longpre, Betty Xiong, Percy Liang, Rishi Bommasani
|
The 2025 Foundation Model Transparency Index
|
Website: https://crfm.stanford.edu/fmti/December-2025/index.html
| null | null | null |
cs.AI cs.CY cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Foundation model developers are among the world's most important companies. As these companies become increasingly consequential, how do their transparency practices evolve? The 2025 Foundation Model Transparency Index is the third edition of an annual effort to characterize and quantify the transparency of foundation model developers. The 2025 FMTI introduces new indicators related to data acquisition, usage data, and monitoring and evaluates companies like Alibaba, DeepSeek, and xAI for the first time. The 2024 FMTI reported that transparency was improving, but the 2025 FMTI finds this progress has deteriorated: the average score out of 100 fell from 58 in 2024 to 40 in 2025. Companies are most opaque about their training data and training compute as well as the post-deployment usage and impact of their flagship models. In spite of this general trend, IBM stands out as a positive outlier, scoring 95, in contrast to the lowest scorers, xAI and Midjourney, at just 14. The five members of the Frontier Model Forum we score end up in the middle of the Index: we posit that these companies avoid reputational harms from low scores but lack incentives to be transparency leaders. As policymakers around the world increasingly mandate certain types of transparency, this work reveals the current state of transparency for foundation model developers, how it may change given newly enacted policy, and where more aggressive policy interventions are necessary to address critical information deficits.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:01:53 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Wan",
"Alexander",
""
],
[
"Klyman",
"Kevin",
""
],
[
"Kapoor",
"Sayash",
""
],
[
"Maslej",
"Nestor",
""
],
[
"Longpre",
"Shayne",
""
],
[
"Xiong",
"Betty",
""
],
[
"Liang",
"Percy",
""
],
[
"Bommasani",
"Rishi",
""
]
] |
2512.10170
|
Lucas Dunker
|
Lucas Dunker, Sai Akshay Menta, Snigdha Mohana Addepalli, Venkata Krishna Rayalu Garapati
|
Semantic-Aware Confidence Calibration for Automated Audio Captioning
|
5 pages, 2 figures
| null | null | null |
cs.SD cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Automated audio captioning models frequently produce overconfident predictions regardless of semantic accuracy, limiting their reliability in deployment. This deficiency stems from two factors: evaluation metrics based on n-gram overlap that fail to capture semantic correctness, and the absence of calibrated confidence estimation. We present a framework that addresses both limitations by integrating confidence prediction into audio captioning and redefining correctness through semantic similarity. Our approach augments a Whisper-based audio captioning model with a learned confidence prediction head that estimates uncertainty from decoder hidden states. We employ CLAP audio-text embeddings and sentence transformer similarities (FENSE) to define semantic correctness, enabling Expected Calibration Error (ECE) computation that reflects true caption quality rather than surface-level text overlap. Experiments on Clotho v2 demonstrate that confidence-guided beam search with semantic evaluation achieves dramatically improved calibration (CLAP-based ECE of 0.071) compared to greedy decoding baselines (ECE of 0.488), while simultaneously improving caption quality across standard metrics. Our results establish that semantic similarity provides a more meaningful foundation for confidence calibration in audio captioning than traditional n-gram metrics.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:09:23 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Dunker",
"Lucas",
""
],
[
"Menta",
"Sai Akshay",
""
],
[
"Addepalli",
"Snigdha Mohana",
""
],
[
"Garapati",
"Venkata Krishna Rayalu",
""
]
] |
2512.10171
|
Jose Hermenegildo Ramirez Gonzalez
|
Gustavo O. Carvalho, F\'abio P. Machado and J. Hermenegildo R. Gonz\'alez
|
The Frog Model on $\mathbb{Z}$ with General Random Survival Parameter
|
20 pages
| null | null | null |
math.PR
|
http://creativecommons.org/licenses/by/4.0/
|
We study the frog model on $\mathbb{Z}$ with particle-wise random geometric lifetimes: each particle has a survival parameter $\pi\in(0,1)$ sampled i.i.d., whose density near $1$ satisfies $f_\pi(u)\sim (1-u)^{\beta-1}L\big((1-u)^{-1}\big)$ with $\beta>0$, and $L$ slowly varying. This strictly extends the $\mathrm{Beta}(\alpha,\beta)$ case. Let $\eta$ denote the common law of the i.i.d.\ initial number of particles $\{\eta_x\}_{x\in\mathbb{Z}}$. Using a percolation comparison and sharp one-particle displacement tails, we obtain a universal threshold at $\beta=\tfrac12$. If $\beta>\tfrac12$ and $E(\eta)<\infty$, extinction occurs almost surely. If $\beta<\tfrac12$ and $\mathbb{P}(\eta=0)<1$, survival has positive probability. At the boundary $\beta=\tfrac12$ we give sharp criteria: extinction if $E(\eta)<\infty$ and $8\,\limsup_{n\to\infty}L(n^2)<1/E(\eta)$; survival if $\mathbb{P}(\eta=0)<1$ and $\sqrt{2}\,\liminf_{n\to\infty}L(n^2)>1/E(\eta)$. These results recover the Carvalho-Machado threshold for Beta laws and show that only the exponent $\beta$ governs the phase transition, while $L$ impacts the critical regime.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:10:50 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Carvalho",
"Gustavo O.",
""
],
[
"Machado",
"Fábio P.",
""
],
[
"González",
"J. Hermenegildo R.",
""
]
] |
2512.10172
|
Nicholas Clark
|
Nicholas Clark, Ryan Bai, Tanu Mitra
|
Offscript: Automated Auditing of Instruction Adherence in LLMs
| null | null | null | null |
cs.HC cs.AI cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Large Language Models (LLMs) and generative search systems are increasingly used for information seeking by diverse populations with varying preferences for knowledge sourcing and presentation. While users can customize LLM behavior through custom instructions and behavioral prompts, no mechanism exists to evaluate whether these instructions are being followed effectively. We present Offscript, an automated auditing tool that efficiently identifies potential instruction following failures in LLMs. In a pilot study analyzing custom instructions sourced from Reddit, Offscript detected potential deviations from instructed behavior in 86.4% of conversations, 22.2% of which were confirmed as material violations through human review. Our findings suggest that automated auditing serves as a viable approach for evaluating compliance to behavioral instructions related to information seeking.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:11:50 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Clark",
"Nicholas",
""
],
[
"Bai",
"Ryan",
""
],
[
"Mitra",
"Tanu",
""
]
] |
2512.10173
|
Stefan Zetzsche
|
Mantas Baksys, Stefan Zetzsche, Olivier Bouissou, Remi Delmas, Soonho Kong
|
ATLAS: Automated Toolkit for Large-Scale Verified Code Synthesis
| null | null | null | null |
cs.SE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Large language models have shown potential for program verification, but progress is hindered by the scarcity of verified code for training. We present ATLAS, an automated pipeline that synthesizes verified programs at scale to address this data bottleneck. ATLAS generates complete Dafny programs with specifications, implementations, and proofs, producing 2.7K verified programs from which we extract over 19K training examples--more than 7 per verified program--by decomposing the synthesis process into multiple specialized tasks. Fine-tuning Qwen 2.5 7B Coder on this dataset produces substantial gains: +23 percentage points on DafnyBench and +50 percentage points on DafnySynthesis. These results demonstrate that synthetic verified code can effectively enhance LLM capabilities for formal verification.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:21:06 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Baksys",
"Mantas",
""
],
[
"Zetzsche",
"Stefan",
""
],
[
"Bouissou",
"Olivier",
""
],
[
"Delmas",
"Remi",
""
],
[
"Kong",
"Soonho",
""
]
] |
2512.10174
|
Andreas Nickl
|
Andreas Nickl, Nard Dumoulin Stuyck, Paul Steinacker, Jesus D. Cifuentes, Santiago Serrano, MengKe Feng, Ensar Vahapoglu, Fay E. Hudson, Kok Wai Chan, Stefan Kubicek, Julien Jussot, Yann Canvel, Sofie Beyne, Yosuke Shimura, Roger Loo, Clement Godfrin, Bart Raes, Sylvain Baudot, Danny Wan, Arne Laucht, Chih-Hwan Yang, Wee Han Lim, Andre Saraiva, Christopher C. Escott, Kristiaan De Greve, Andrew S. Dzurak, Tuomo Tanttu
|
Eight-Qubit Operation of a 300 mm SiMOS Foundry-Fabricated Device
| null | null | null | null |
quant-ph
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Silicon spin qubits are a promising candidate for quantum computing, thanks to their high coherence, high controllability and manufacturability. However, the most scalable complementary metal-oxide-semiconductor (CMOS) based implementations have so far been limited to a few qubits. Here, to take a step towards large scale systems, we tune and coherently control an eight-dot linear array of silicon spin qubits fabricated in 300 mm CMOS-compatible foundry process, establishing operational scalability beyond the two-qubit regime. All eight qubits are successfully tuned and characterized as four double dot pairs, exhibiting Ramsey dephasing times $T_2^*$ up to 41(2) $\mu$s and Hahn-echo coherence times $T_2^{\mathrm{Hahn}}$ up to 1.31(4) ms. Readout of the central four qubits is achieved via a cascaded charge-sensing protocol, enabling simultaneous high-fidelity measurements of the entire multi-qubit array. Additionally, we demonstrate a two-qubit gate operation between adjacent qubits with low phase noise. We demonstrate here that we can scale silicon spin qubit arrays to medium-sized arrays of 8 qubits while maintaining coherence of the system.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:21:16 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Nickl",
"Andreas",
""
],
[
"Stuyck",
"Nard Dumoulin",
""
],
[
"Steinacker",
"Paul",
""
],
[
"Cifuentes",
"Jesus D.",
""
],
[
"Serrano",
"Santiago",
""
],
[
"Feng",
"MengKe",
""
],
[
"Vahapoglu",
"Ensar",
""
],
[
"Hudson",
"Fay E.",
""
],
[
"Chan",
"Kok Wai",
""
],
[
"Kubicek",
"Stefan",
""
],
[
"Jussot",
"Julien",
""
],
[
"Canvel",
"Yann",
""
],
[
"Beyne",
"Sofie",
""
],
[
"Shimura",
"Yosuke",
""
],
[
"Loo",
"Roger",
""
],
[
"Godfrin",
"Clement",
""
],
[
"Raes",
"Bart",
""
],
[
"Baudot",
"Sylvain",
""
],
[
"Wan",
"Danny",
""
],
[
"Laucht",
"Arne",
""
],
[
"Yang",
"Chih-Hwan",
""
],
[
"Lim",
"Wee Han",
""
],
[
"Saraiva",
"Andre",
""
],
[
"Escott",
"Christopher C.",
""
],
[
"De Greve",
"Kristiaan",
""
],
[
"Dzurak",
"Andrew S.",
""
],
[
"Tanttu",
"Tuomo",
""
]
] |
2512.10175
|
Seog-Jin Kim
|
Seog-Jin Kim and Rong Luo
|
Squares of subcubic planar graphs without cycles of length 4-8 are 6-choosable
| null | null | null | null |
math.CO
|
http://creativecommons.org/licenses/by/4.0/
|
The {\em square} of a graph $G$, denoted $G^2$, has the same vertex set as $G$ and an edge between any two vertices at distance at most $2$ in $G$. Wegner (1977) conjectured that for a planar graph $G$, $\chi(G^2) \leq 7$ if $\Delta(G) = 3$, $\chi(G^2) \leq \Delta(G)+5$ if $4 \leq \Delta(G) \leq 7$, and $\chi(G^2) \leq \lfloor 3\Delta(G)/2 \rfloor$ if $\Delta(G) \geq 8$, and Thomassen (2018) confirmed the conjecture for $\Delta(G) = 3$. Dvo\v{r}\'{a}k et al. (2008) and Feder et al. (2021) further conjectured that $\chi(G^2) \leq 6$ for cubic bipartite planar graphs. A natural question is whether this bound also holds for the list-chromatic number, i.e., whether $\chi_{\ell}(G^2) \leq 6$ for such graphs. More generally, it is of interest to determine sufficient conditions ensuring $\chi_{\ell}(G^2) \leq 6$ for subcubic planar graphs. In this paper, we prove that $\chi_{\ell}(G^2) \leq 6$ for subcubic planar graphs containing no $k$-cycles for $4 \leq k \leq 8$, improving a result of Cranston and Kim (2008).
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:23:53 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Kim",
"Seog-Jin",
""
],
[
"Luo",
"Rong",
""
]
] |
2512.10176
|
Shinya Sugiura
|
Phuc V. Trinh, Shinya Sugiura, Carlo Ottaviani, Chao Xu, Lajos Hanzo
|
Simultaneous Classical and Quantum Communications: Recent Progress and Three Challenges
|
8 pages, 5 figures, 1 table
| null |
10.1109/MNET.2025.3643793
| null |
quant-ph eess.SP
|
http://creativecommons.org/licenses/by/4.0/
|
A critical aspect of next-generation wireless networks is the integration of quantum communications to guard against quantum computing threats to classical networks. Despite successful experimental demonstrations, integrating quantum communications into the classical infrastructure faces substantial challenges, including high costs, compatibility issues, and extra hardware deployment to accommodate both classical and quantum communication equipment. To mitigate these challenges, we explore novel protocols that enable simultaneous classical and quantum communications, relying on a single set of transceivers to jointly modulate and decode classical and quantum information onto the same signal. Additionally, we emphasize extending quantum communication capabilities beyond traditional optical bands into the terahertz, even possibly to millimeter-wave and microwave frequencies, thereby broadening the potential horizon of quantum-secure applications. Finally, we identify open problems that must be addressed to facilitate practical implementation.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:27:08 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Trinh",
"Phuc V.",
""
],
[
"Sugiura",
"Shinya",
""
],
[
"Ottaviani",
"Carlo",
""
],
[
"Xu",
"Chao",
""
],
[
"Hanzo",
"Lajos",
""
]
] |
2512.10177
|
Simon MacLean
|
Shamil Asgarli, Sara Krehbiel, Simon MacLean
|
Bell coloring graphs: realizability and reconstruction
|
31 pages, 15 figures
| null | null | null |
math.CO
|
http://creativecommons.org/licenses/by/4.0/
|
Given a graph $G$, the Bell $k$-coloring graph $\mathcal{B}_k(G)$ has vertices given by partitions of $V(G)$ into $k$ independent sets (allowing empty parts), with two partitions adjacent if they differ only in the placement of a single vertex. We first give a structural classification of cliques in Bell coloring graphs. We then show that all trees and all cycles arise as Bell coloring graphs, while $K_4-e$ is not a Bell coloring graph and, more generally, $K_n-e$ is not an induced subgraph of any Bell coloring graph whenever $n \geq 6$. We also prove two reconstruction results: the Bell $3$-coloring graph is a complete invariant for trees, and the Bell $n$-coloring multigraph determines any graph up to universal vertices.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:31:56 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Asgarli",
"Shamil",
""
],
[
"Krehbiel",
"Sara",
""
],
[
"MacLean",
"Simon",
""
]
] |
2512.10178
|
Keito Inoshita
|
Keito Inoshita, Xiaokang Zhou, Akira Kawai, Katsutoshi Yada
|
CIEGAD: Cluster-Conditioned Interpolative and Extrapolative Framework for Geometry-Aware and Domain-Aligned Data Augmentation
| null | null | null | null |
cs.LG cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In practical deep learning deployment, the scarcity of data and the imbalance of label distributions often lead to semantically uncovered regions within the real-world data distribution, hindering model training and causing misclassification near class boundaries as well as unstable behaviors in peripheral areas. Although recent large language models (LLMs) show promise for data augmentation, an integrated framework that simultaneously achieves directional control of generation, domain alignment, and quality control has not yet been fully established. To address these challenges, we propose a Cluster-conditioned Interpolative and Extrapolative framework for Geometry-Aware and Domain-aligned data augmentation (CIEGAD), which systematically complements both in-distribution and out-of-distribution semantically uncovered regions. CIEGAD constructs domain profiles through cluster conditioning, allocates generation with a hierarchical frequency-geometric allocation integrating class frequency and geometric indicators, and finely controls generation directions via the coexistence of interpolative and extrapolative synthesis. It further performs quality control through geometry-constrained filtering combined with an LLM-as-a-Judge mechanism. Experiments on multiple classification tasks demonstrate that CIEGAD effectively extends the periphery of real-world data distributions while maintaining high alignment between generated and real-world data as well as semantic diversity. In particular, for long-tailed and multi-class classification tasks, CIEGAD consistently improves F1 and recall, validating the triple harmony of distributional consistency, diversity, and quality. These results indicate that CIEGAD serves as a practically oriented data augmentation framework that complements underrepresented regions while preserving alignment with real-world data.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:32:37 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Inoshita",
"Keito",
""
],
[
"Zhou",
"Xiaokang",
""
],
[
"Kawai",
"Akira",
""
],
[
"Yada",
"Katsutoshi",
""
]
] |
2512.10179
|
Abolfazl Shahrooei
|
Abolfazl Shahrooei, Luke Arthur, Om Patel, and Derek Kamper
|
Assessing Neuromorphic Computing for Fingertip Force Decoding from Electromyography
|
5 pages, 6 figures. Poster included as ancillary file (IEEE_NER2025_NeuromorphicEMG_poster.pdf). Presented at IEEE EMBS NER 2025, also at NC State College of Engineering Applied AI Symposium and NC State ECE Graduate Research Symposium (tied for Best Poster)
| null | null | null |
cs.LG eess.SP
|
http://creativecommons.org/licenses/by/4.0/
|
High-density surface electromyography (HD-sEMG) provides a noninvasive neural interface for assistive and rehabilitation control, but mapping neural activity to user motor intent remains challenging. We assess a spiking neural network (SNN) as a neuromorphic architecture against a temporal convolutional network (TCN) for decoding fingertip force from motor-unit (MU) firing derived from HD-sEMG. Data were collected from a single participant (10 trials) with two forearm electrode arrays; MU activity was obtained via FastICA-based decomposition, and models were trained on overlapping windows with end-to-end causal convolutions. On held-out trials, the TCN achieved 4.44% MVC RMSE (Pearson r = 0.974) while the SNN achieved 8.25% MVC (r = 0.922). While the TCN was more accurate, we view the SNN as a realistic neuromorphic baseline that could close much of this gap with modest architectural and hyperparameter refinements.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:33:31 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Shahrooei",
"Abolfazl",
""
],
[
"Arthur",
"Luke",
""
],
[
"Patel",
"Om",
""
],
[
"Kamper",
"Derek",
""
]
] |
2512.10180
|
Pracheta Harlikar
|
Pracheta Harlikar, Abdel-Hameed A. Badawy, Prasanna Date
|
Neuromorphic Processor Employing FPGA Technology with Universal Interconnections
| null | null | null | null |
cs.AR
|
http://creativecommons.org/licenses/by/4.0/
|
Neuromorphic computing, inspired by biological neural systems, holds immense promise for ultra-low-power and real-time inference applications. However, limited access to flexible, open-source platforms continues to hinder widespread adoption and experimentation. In this paper, we present a low-cost neuromorphic processor implemented on a Xilinx Zynq-7000 FPGA platform. The processor supports all-to-all configurable connectivity and employs the leaky integrate-and-fire (LIF) neuron model with customizable parameters such as threshold, synaptic weights, and refractory period. Communication with the host system is handled via a UART interface, enabling runtime reconfiguration without hardware resynthesis. The architecture was validated using benchmark datasets including the Iris classification and MNIST digit recognition tasks. Post-synthesis results highlight the design's energy efficiency and scalability, establishing its viability as a research-grade neuromorphic platform that is both accessible and adaptable for real-world spiking neural network applications. This implementation will be released as open source following project completion.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:35:48 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Harlikar",
"Pracheta",
""
],
[
"Badawy",
"Abdel-Hameed A.",
""
],
[
"Date",
"Prasanna",
""
]
] |
2512.10181
|
Shinya Sugiura
|
Phuc V. Trinh, Shinya Sugiura
|
Quantum Internet in the Sky: Vision, Challenges, Solutions, and Future Directions
|
7 pages, 4 figures, 1 table
|
IEEE Communications Magazine, vol. 62, no. 10, pp. 62-68, Oct. 2024
|
10.1109/MCOM.003.2300835
| null |
quant-ph eess.SP
|
http://creativecommons.org/licenses/by/4.0/
|
This article envisions the concept of a ``Quantum Internet in the Sky", aiming to establish ubiquitous quantum communication links among distant nodes via free-space optical channels. Our key focus is on deploying quantum communication terminals on non-terrestrial platforms, specifically unmanned aerial vehicles and satellites, at various altitudes. By highlighting the unique characteristics of these platforms compared to terrestrial counterparts, we address inherent challenges and discuss potential solutions through meticulous system designs and analyses of typical non-terrestrial quantum communication scenarios. Finally, we illuminate the path forward by proposing essential future directions that underscore the integration of high-dimensional multipartite quantum communications with sensing, computing, and intelligence for multiple users en route to realizing a fully operational Quantum Internet.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:36:36 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Trinh",
"Phuc V.",
""
],
[
"Sugiura",
"Shinya",
""
]
] |
2512.10182
|
Daisuke Kishimoto
|
Tsuyoshi Kato, Daisuke Kishimoto, Mitsunobu Tsutaya
|
Uniform Lefschetz fixed-point theory
|
37 pages
| null | null | null |
math.AT math.GT
|
http://creativecommons.org/licenses/by/4.0/
|
We develop the Lefschetz fixed-point theory for noncompact manifolds of bounded geometry and uniformly continuous maps. Specifically, we define the uniform Lefschetz class $\mathscr{L}(f)$ of a uniformly continuous map $f\colon M\to M$ of a uniform simply-connected noncompact complete Riemannian manifold of bounded geometry $M$ satisfying $d(f,1)<\infty$, and prove that $\mathscr{L}(f)=0$ if and only if $f$ is uniformly homotopic to a strongly fixed-point free (without fixed-points on $M$ and at infinity) uniformly continuous map. To achieve this, we introduce a new cohomology for metric spaces, called uniform bounded cohomology, which is a variant of bounded cohomology, and develop an obstruction theory formulated in terms of this cohomology.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:42:28 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Kato",
"Tsuyoshi",
""
],
[
"Kishimoto",
"Daisuke",
""
],
[
"Tsutaya",
"Mitsunobu",
""
]
] |
2512.10183
|
Gonzalo Mateos
|
Gonzalo Mateos, Yanning Shen, Georgios B. Giannakis, Ananthram Swami
|
Topology Identification and Inference over Graphs
|
Contributed chapter to appear in Handbook of Statistics Volume 54: Multidimensional Signal Processing, K. V. Mishra, G. R. Arce, and A. S. R. S. Rao, Editors, Amsterdam, Netherlands, Elsevier, 2026
| null | null | null |
eess.SP cs.SI stat.ME stat.ML
|
http://creativecommons.org/licenses/by/4.0/
|
Topology identification and inference of processes evolving over graphs arise in timely applications involving brain, transportation, financial, power, as well as social and information networks. This chapter provides an overview of graph topology identification and statistical inference methods for multidimensional relational data. Approaches for undirected links connecting graph nodes are outlined, going all the way from correlation metrics to covariance selection, and revealing ties with smooth signal priors. To account for directional (possibly causal) relations among nodal variables and address the limitations of linear time-invariant models in handling dynamic as well as nonlinear dependencies, a principled framework is surveyed to capture these complexities through judiciously selected kernels from a prescribed dictionary. Generalizations are also described via structural equations and vector autoregressions that can exploit attributes such as low rank, sparsity, acyclicity, and smoothness to model dynamic processes over possibly time-evolving topologies. It is argued that this approach supports both batch and online learning algorithms with convergence rate guarantees, is amenable to tensor (that is, multi-way array) formulations as well as decompositions that are well-suited for multidimensional network data, and can seamlessly leverage high-order statistical information.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:47:09 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Mateos",
"Gonzalo",
""
],
[
"Shen",
"Yanning",
""
],
[
"Giannakis",
"Georgios B.",
""
],
[
"Swami",
"Ananthram",
""
]
] |
2512.10184
|
Koji Ishiwata
|
Koji Ishiwata, Eiichiro Komatsu
|
Perturbative limits on axion-SU(2) gauge dynamics during inflation from the energy density of spin-2 particles
|
26 pages, 11 figures
| null | null |
KANAZAWA-25-05
|
hep-ph
|
http://creativecommons.org/licenses/by/4.0/
|
We investigate the conditions under which the perturbative treatment of the backreaction of spin-2 particles on the dynamics of an axion-SU(2) gauge field system breaks down during cosmic inflation. This condition is based on the ratio of the energy density of spin-2 particles from the SU(2) gauge field to that of the background field. The perturbative treatment breaks down when this ratio exceeds unity. We show that this occurs within a parameter space nearly identical to the strong backreaction regime identified in previous studies. However, in some cases, the ratio exceeds unity even before the system enters the strong backreaction regime. Our results suggest that attempts to study the strong backreaction regime using perturbation theory are necessarily limited. Reliable calculations require non-perturbative treatments, such as three-dimensional lattice simulations.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:48:21 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ishiwata",
"Koji",
""
],
[
"Komatsu",
"Eiichiro",
""
]
] |
2512.10185
|
Yangkun Wang
|
Yangkun Wang, Jingbo Shang
|
Watermarks for Language Models via Probabilistic Automata
| null | null | null | null |
cs.CR cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
A recent watermarking scheme for language models achieves distortion-free embedding and robustness to edit-distance attacks. However, it suffers from limited generation diversity and high detection overhead. In parallel, recent research has focused on undetectability, a property ensuring that watermarks remain difficult for adversaries to detect and spoof. In this work, we introduce a new class of watermarking schemes constructed through probabilistic automata. We present two instantiations: (i) a practical scheme with exponential generation diversity and computational efficiency, and (ii) a theoretical construction with formal undetectability guarantees under cryptographic assumptions. Extensive experiments on LLaMA-3B and Mistral-7B validate the superior performance of our scheme in terms of robustness and efficiency.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:49:06 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Wang",
"Yangkun",
""
],
[
"Shang",
"Jingbo",
""
]
] |
2512.10186
|
Matthew A. Malkan
|
Matthew Malkan, Lisbeth Jensen, and Lei Hao
|
The Narrow Emission Lines of Seyfert 1 Galaxies: Comparisons with a Large SDSS Sample
|
31 pages, 12 figures
| null | null | null |
astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
We analyzed a large sample of SDSS spectra of Seyfert galaxies, subdividing Seyfert 1s based on their narrow-to-broad Halpha components. Comparing their narrow-lines (NL) to Seyfert2s in line-ratio diagrams, most of the NL of strong Sy 1.0 and Sy 1.2s (with dominant broad lines) are the same as those of pure Sy2s. In contrast, only 25-30 percent of the Sy1.8 and Sy1.9 nuclei (with weak broad lines) are located in the pure Sy2 region, with the rest falling in the composite-star-forming region. We explain these Seyfert-plus-star-formation spectra with a simple model. It shows that 85 percent of NL in Sy1.9 are from HII-regions, while 88 percent of the NL in Sy 1.0 arise from the same NLR as in pure Sy2. About 6 percent of the strong and weak Sy1's have NL dominated by LINER emission, while 15 percent of intermediate Seyferts (Sy 1.5 and Sy 1.6) do. To confirm this Seyfert 1 AGN plus star formation combination, we used stellar absorption-lines to compare their stellar populations. Their Hdelta strengths show that LINERs, pure Sy2s, and also the broad-line dominated Sy1s have old stellar populations. The weak Sy 1s show stronger Hdelta absorption, indicating larger proportions of young stars. About one third of the u band light in Sy1.0 and 1.2 is blended Balmer lines and continuum from the BLR. The NL gas reddening increases as the BLR strength decreases, from Sy1.0 (0.13 mag), to Sy1.9 (0.40 mag), to Sy2s and LINERs both with 0.50 mag. Our data do not support the simplest version of Seyfert 1 and 2 unification, where both AGN classes have identical NL.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:49:57 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Malkan",
"Matthew",
""
],
[
"Jensen",
"Lisbeth",
""
],
[
"Hao",
"Lei",
""
]
] |
2512.10187
|
Stefan Zetzsche
|
Mantas Baksys, Stefan Zetzsche, Olivier Bouissou
|
MiniF2F-Dafny: LLM-Guided Mathematical Theorem Proving via Auto-Active Verification
| null | null | null | null |
cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present miniF2F-Dafny, the first translation of the mathematical reasoning benchmark miniF2F to an automated theorem prover: Dafny. Previously, the benchmark existed only in interactive theorem provers (Lean, Isabelle, HOL Light, Metamath). We find that Dafny's automation verifies 99/244 (40.6%) of the test set and 109/244 (44.7%) of the validation set with empty proofs--requiring no manual proof steps. For problems where empty proofs fail, we evaluate 12 off-the-shelf LLMs on providing proof hints. The best model we test achieves 55.7% pass@4 success rate employing iterative error correction. These preliminary results highlight an effective division of labor: LLMs provide high-level guidance while automation handles low-level details. Our benchmark can be found on GitHub at http://github.com/dafny-lang/miniF2F .
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:52:19 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Baksys",
"Mantas",
""
],
[
"Zetzsche",
"Stefan",
""
],
[
"Bouissou",
"Olivier",
""
]
] |
2512.10188
|
Gabriel Clara
|
Gabriel Clara and Yazan Mash'al
|
The Interplay of Statistics and Noisy Optimization: Learning Linear Predictors with Random Data Weights
| null | null | null | null |
stat.ML cs.LG stat.CO
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
We analyze gradient descent with randomly weighted data points in a linear regression model, under a generic weighting distribution. This includes various forms of stochastic gradient descent, importance sampling, but also extends to weighting distributions with arbitrary continuous values, thereby providing a unified framework to analyze the impact of various kinds of noise on the training trajectory. We characterize the implicit regularization induced through the random weighting, connect it with weighted linear regression, and derive non-asymptotic bounds for convergence in first and second moments. Leveraging geometric moment contraction, we also investigate the stationary distribution induced by the added noise. Based on these results, we discuss how specific choices of weighting distribution influence both the underlying optimization problem and statistical properties of the resulting estimator, as well as some examples for which weightings that lead to fast convergence cause bad statistical performance.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 00:55:29 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Clara",
"Gabriel",
""
],
[
"Mash'al",
"Yazan",
""
]
] |
2512.10189
|
Hengameh R. Dehkordi
|
Hengameh R. Dehkordi
|
Wildfire Propagation Modeling using Satellite-Derived Parameters and Generalized Elliptical Frames
|
14 figures
| null | null | null |
math-ph math.MP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Wildfires pose significant threats to ecosystems and communities, yet accurately modeling fire spread remains challenging, particularly in regions where environmental and fuel data are scarce or unavailable. This study introduces an innovative conceptual and methodological framework for simulating wildfire propagation and estimating the rate of spread using a hybrid geometric and data-driven approach that relies exclusively on multi-source satellite observations. The framework integrates thermal fire-front detections, atmospheric conditions, and vegetation indices using two complementary geometric modeling strategies. The first strategy applies the Huygens principle, where generalized elliptical frames are expanded locally at every point along the fire perimeter, and their combined envelope forms the evolving wavefront. This method is best suited for situations in which environmental variables are available and can be incorporated to refine the anisotropic spread function. The second strategy relies solely on the generalized elliptical frames themselves; for each time step, an elliptical frame is constructed from the inferred head and back rates of spread and wind, and the burned area is obtained by enclosing the region determined by these curves. Together, these two methods provide a flexible toolkit that adapts to both data-rich and data-limited conditions while retaining a unified geometric interpretation of wildfire spread. To demonstrate the applicability of the method, we present a case study based on the Eaton Fire, January 2025, using publicly available multi-day satellite imagery. Despite the absence of complete operational datasets for that event, the model driven only by satellite-derived parameters reproduces key qualitative features of the observed propagation pattern, underscoring the flexibility and robustness of the proposed approach in data-limited contexts.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 01:01:07 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Dehkordi",
"Hengameh R.",
""
]
] |
2512.10190
|
Xizhi Liu
|
Xizhi Liu, Sijie Ren, and Jian Wang
|
Andr{\'a}sfai--Erd\H{o}s--S\'{o}s theorem under max-degree constraints
|
12pp, comments are welcome
| null | null | null |
math.CO
|
http://creativecommons.org/licenses/by/4.0/
|
We establish the following strengthening of the celebrated Andr{\'a}sfai--Erd\H{o}s--S\'{o}s theorem: If $G$ is an $n$-vertex $K_{r+1}$-free graph whose minimum degree $\delta(G)$ and maximum degree $\Delta(G)$ satisfy
\begin{align*}
\delta(G) > \min \left\{ \frac{3r-4}{3r-2}n-\frac{\Delta(G)}{3r-2},~n-\frac{\Delta(G)+1}{r-1} \right\},
\end{align*}
then $G$ is $r$-partite. This bound is tight for all feasible values of $\Delta(G)$. We also obtain an analogous tight result for graphs with large odd girth.
Our proof does not rely on the Andr{\'a}sfai--Erd\H{o}s--S\'{o}s theorem itself, and therefore yields an alternative proof of this classical result.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 01:03:26 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Liu",
"Xizhi",
""
],
[
"Ren",
"Sijie",
""
],
[
"Wang",
"Jian",
""
]
] |
2512.10191
|
Hao Shu
|
Hao Shu, Jicheng Li, Yu Jin, and Ling Zhou
|
Exact Recovery of Non-Random Missing Multidimensional Time Series via Temporal Isometric Delay-Embedding Transform
| null | null | null | null |
cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Non-random missing data is a ubiquitous yet undertreated flaw in multidimensional time series, fundamentally threatening the reliability of data-driven analysis and decision-making. Pure low-rank tensor completion, as a classical data recovery method, falls short in handling non-random missingness, both methodologically and theoretically. Hankel-structured tensor completion models provide a feasible approach for recovering multidimensional time series with non-random missing patterns. However, most Hankel-based multidimensional data recovery methods both suffer from unclear sources of Hankel tensor low-rankness and lack an exact recovery theory for non-random missing data. To address these issues, we propose the temporal isometric delay-embedding transform, which constructs a Hankel tensor whose low-rankness is naturally induced by the smoothness and periodicity of the underlying time series. Leveraging this property, we develop the \textit{Low-Rank Tensor Completion with Temporal Isometric Delay-embedding Transform} (LRTC-TIDT) model, which characterizes the low-rank structure under the \textit{Tensor Singular Value Decomposition} (t-SVD) framework. Once the prescribed non-random sampling conditions and mild incoherence assumptions are satisfied, the proposed LRTC-TIDT model achieves exact recovery, as confirmed by simulation experiments under various non-random missing patterns. Furthermore, LRTC-TIDT consistently outperforms existing tensor-based methods across multiple real-world tasks, including network flow reconstruction, urban traffic estimation, and temperature field prediction. Our implementation is publicly available at https://github.com/HaoShu2000/LRTC-TIDT.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 01:04:27 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Shu",
"Hao",
""
],
[
"Li",
"Jicheng",
""
],
[
"Jin",
"Yu",
""
],
[
"Zhou",
"Ling",
""
]
] |
2512.10192
|
Michele Botti
|
Stefano Bonetti, Michele Botti, and Patrick Vega
|
A robust fully-mixed finite element method with skew-symmetry penalization for low-frequency poroelasticity
|
19 pages
| null | null | null |
math.NA cs.NA
|
http://creativecommons.org/licenses/by/4.0/
|
In this work, we present and analyze a fully-mixed finite element scheme for the dynamic poroelasticity problem in the low-frequency regime. We write the problem as a four-field, first-order, hyperbolic system of equations where the symmetry constraint on the stress field is imposed via penalization. This strategy is equivalent to adding a perturbation to the saddle point system arising when the stress symmetry is weakly-imposed. The coupling of solid and fluid phases is discretized by means of stable mixed elements in space and implicit time advancing schemes. The presented stability analysis is fully robust with respect to meaningful cases of degenerate model parameters. Numerical tests validate the convergence and robustness and assess the performances of the method for the simulation of wave propagation phenomena in porous materials.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 01:11:32 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Bonetti",
"Stefano",
""
],
[
"Botti",
"Michele",
""
],
[
"Vega",
"Patrick",
""
]
] |
2512.10193
|
Chi Kin Tam
|
C.K. Tam, Z. Chaj\k, R.S. Wang, F.C.E. Teh, N. Ikeno, W.G. Lynch, A. Ono, M.B. Tsang, A. Anthony, S. Barlini, J. Barney, K.W. Brown, A. Camaiani, A. Chbihi, D. Dell'Aquila, J. Estee, A. Galindo-Uribarri, F. Guan, B. Hong, T. Isobe, G. Jhang, O.B. Khanal, Y.J. Kim, H.S. Lee, J.W. Lee, J.-W. Lee, J. Manfredi, L. Morelli, P. Morfouace, S.H. Nam, C.Y. Niu, E. Padilla-Rodal, J. Park, S. Sweany, C.Y. Tsang, G. Verde, J. Wieske, K. Zhu
|
Impact of the in-medium cross section on cluster spectra in ${}^{40,48}\mathrm{Ca}+{}^{58,64}\mathrm{Ni}$ collisions at $56$ and $140$ $\mathbf{\mathrm{MeV}}/\mathrm{\mathbf{nucleon}}$
| null | null | null | null |
nucl-th nucl-ex
|
http://creativecommons.org/licenses/by/4.0/
|
Although significant efforts have been made to investigate the density dependence of the nuclear symmetry energy, the influence of the in-medium cross section on particle production in transport models is not well constrained. The in-medium cross section reflects the dynamic situation of the medium such as a nontrivial phase space distribution. In this study, we analyze the transverse momentum spectra of $p$, $d$, $t$, ${}^3{\mathrm{He}}$ and $\alpha$ particles emitted near mid-rapidity in central $^{40,48}\mathrm{Ca}$ + $^{58, 64}\mathrm{Ni}$ reactions at $56$ and $140$ $\mathrm{MeV}/\mathrm{nucleon}$. The Antisymmetrized Molecular Dynamics ($\mathrm{AMD}$) model is chosen as the transport model for data comparison. Central events are selected based on charged-particle multiplicity in both the experimental data and AMD calculations after applying an experimental filter. Our results show that the in-medium nucleon-nucleon scattering cross-sections are more strongly reduced at $56$ $\mathrm{MeV}/\mathrm{nucleon}$ than at $140$ $\mathrm{MeV}/\mathrm{nucleon}$ incident energy.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 01:13:05 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Tam",
"C. K.",
""
],
[
"Chaj\\k",
"Z.",
""
],
[
"Wang",
"R. S.",
""
],
[
"Teh",
"F. C. E.",
""
],
[
"Ikeno",
"N.",
""
],
[
"Lynch",
"W. G.",
""
],
[
"Ono",
"A.",
""
],
[
"Tsang",
"M. B.",
""
],
[
"Anthony",
"A.",
""
],
[
"Barlini",
"S.",
""
],
[
"Barney",
"J.",
""
],
[
"Brown",
"K. W.",
""
],
[
"Camaiani",
"A.",
""
],
[
"Chbihi",
"A.",
""
],
[
"Dell'Aquila",
"D.",
""
],
[
"Estee",
"J.",
""
],
[
"Galindo-Uribarri",
"A.",
""
],
[
"Guan",
"F.",
""
],
[
"Hong",
"B.",
""
],
[
"Isobe",
"T.",
""
],
[
"Jhang",
"G.",
""
],
[
"Khanal",
"O. B.",
""
],
[
"Kim",
"Y. J.",
""
],
[
"Lee",
"H. S.",
""
],
[
"Lee",
"J. W.",
""
],
[
"Lee",
"J. -W.",
""
],
[
"Manfredi",
"J.",
""
],
[
"Morelli",
"L.",
""
],
[
"Morfouace",
"P.",
""
],
[
"Nam",
"S. H.",
""
],
[
"Niu",
"C. Y.",
""
],
[
"Padilla-Rodal",
"E.",
""
],
[
"Park",
"J.",
""
],
[
"Sweany",
"S.",
""
],
[
"Tsang",
"C. Y.",
""
],
[
"Verde",
"G.",
""
],
[
"Wieske",
"J.",
""
],
[
"Zhu",
"K.",
""
]
] |
2512.10194
|
Ruolin Li
|
Jiaxin Hou, Kexin Wang, Ruolin Li, Jong-shi Pang
|
Traffic Equilibrium in Mixed-Autonomy Network with Capped Customer Waiting
|
under review for journal
| null | null | null |
eess.SY cs.SY
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper develops a unified modeling framework to capture the equilibrium-state interactions among ride-hailing companies, travelers, and traffic of mixed-autonomy transportation networks. Our framework integrates four interrelated sub-modules: (i) the operational behavior of representative ride-hailing Mixed-Fleet Traffic Network Companies (MiFleet TNCs) managing autonomous vehicle (AV) and human-driven vehicle (HV) fleets, (ii) traveler mode-choice decisions taking into account travel costs and waiting time, (iii) capped customer waiting times to reflect the option available to travelers not to wait for TNCs' service beyond his/her patience and to resort to existing travel modes, and (iv) a flow-dependent traffic congestion model for travel times. A key modeling feature distinguishes AVs and HVs across the pickup and service (customer-on-board) stages: AVs follow Wardrop pickup routes but may deviate during service under company coordination, whereas HVs operate in the reverse manner. The overall framework is formulated as a Nonlinear Complementarity Problem (NCP), which is equivalent to a Variational Inequality(VI) formulation based on which the existence of a variational equilibrium solution to the traffic model is established. Numerical experiments examine how AV penetration and Wardrop relaxation factors, which bound route deviation, affect company, traveler, and system performance to various degrees. The results provide actionable insights for policymakers on regulating AV adoption and company vehicle deviation behavior in modern-day traffic systems that are fast changing due to the advances in technology and information accessibility.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 01:24:15 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Hou",
"Jiaxin",
""
],
[
"Wang",
"Kexin",
""
],
[
"Li",
"Ruolin",
""
],
[
"Pang",
"Jong-shi",
""
]
] |
2512.10195
|
Byung-Hoon Kim M.D. Ph.D.
|
Gyutaek Oh, Sangjoon Park, Byung-Hoon Kim
|
AutoMedic: An Automated Evaluation Framework for Clinical Conversational Agents with Medical Dataset Grounding
| null | null | null | null |
cs.CL cs.LG cs.MA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Evaluating large language models (LLMs) has recently emerged as a critical issue for safe and trustworthy application of LLMs in the medical domain. Although a variety of static medical question-answering (QA) benchmarks have been proposed, many aspects remain underexplored, such as the effectiveness of LLMs in generating responses in dynamic, interactive clinical multi-turn conversation situations and the identification of multi-faceted evaluation strategies beyond simple accuracy. However, formally evaluating a dynamic, interactive clinical situation is hindered by its vast combinatorial space of possible patient states and interaction trajectories, making it difficult to standardize and quantitatively measure such scenarios. Here, we introduce AutoMedic, a multi-agent simulation framework that enables automated evaluation of LLMs as clinical conversational agents. AutoMedic transforms off-the-shelf static QA datasets into virtual patient profiles, enabling realistic and clinically grounded multi-turn clinical dialogues between LLM agents. The performance of various clinical conversational agents is then assessed based on our CARE metric, which provides a multi-faceted evaluation standard of clinical conversational accuracy, efficiency/strategy, empathy, and robustness. Our findings, validated by human experts, demonstrate the validity of AutoMedic as an automated evaluation framework for clinical conversational agents, offering practical guidelines for the effective development of LLMs in conversational medical applications.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 01:25:36 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Oh",
"Gyutaek",
""
],
[
"Park",
"Sangjoon",
""
],
[
"Kim",
"Byung-Hoon",
""
]
] |
2512.10196
|
Isaac Cho
|
Isaac Cho, Zachary Wartell
|
HyFinBall: a Hybrid User Interface for Coordinated 2D+3D Visualization in Semi-Immersive VR
| null | null | null | null |
cs.HC
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Sophisticated 3D visualization applications usually provide coordinated 2D and 3D views. Normally 3D input device is used for 3D tasks since they perform better than traditional 2D input devices. However, they do not perform better for 2D tasks. This paper presents a bimanual hybrid user interface that supports four interaction modes: a dual 6-degree-of-freedom (DOF) input device mode, a dual planar constrained 3DOF input device mode, a dual 2-finger multi-touch mode, and 3D hand and finger gestures. The application is a multi-dimensional visualization with coordinated 3D and 2D views on a desktop VR system. The input devices are buttonballs with seamless switching between 3D and 2D device modes, as well as between free-hand finger input and device usage. The 3D and 2D device mode switch automatically switches a buttonball's visual representation between a 3D cursor and a 2D cursor while changing the available user interaction techniques between 3D and 2D interaction techniques to interact with the coordinated views. The paper also provides two formal user studies to evaluate HyFinBall for various dimensional tasks, including 3D, 2D, and cross-dimensional tasks. Our experimental results show the benefits of the HyFinBall interface for cross-dimensional tasks that require 3D and 2D interactions.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 01:28:04 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Cho",
"Isaac",
""
],
[
"Wartell",
"Zachary",
""
]
] |
2512.10197
|
Zhen-Hua Zhao
|
Shu-Cong Liu, Yong-Qiang Wang, Zhen-Hua Zhao
|
Frozen solitonic Hayward-boson stars in Anti-de Sitter Spacetime
|
25 pages, 15 figures
| null | null | null |
gr-qc
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We construct solitonic Hayward-boson stars (SHBSs) in Anti-de Sitter (AdS) spacetime, which consists of the Einstein-Hayward model and a complex scalar field with a soliton potential. Our results reveal a critical magnetic charge $q_c$. For $q\geq q_c$ in the limit of $\omega \rightarrow 0$, the matter field is primarily distributed within the critical radius $r_c$, beyond which it decays rapidly, while the metric components $-g_{tt}$ and $1/g_{rr}$ become very small at $r_c$. These solutions are termed ``frozen solitonic Hayward-boson stars" (FSHBSs). Continuously decreasing $\Lambda$ disrupts the frozen state. However, we did not find a frozen solution when $q<q_c$. The value of $q_c$ depends both on the cosmological constant $\Lambda$ and the self-interaction coupling $\eta$. We also found that for high frequency solutions, increasing $\eta$ can yield a pure Hayward solution. However, for low frequency solutions, increasing $\eta$ reduces both $1/g_{rr}$ and $-g_{tt}$. Furthermore, we analyzed the effective potential of SHBSs and identified an extra pair of light rings in the second solution branch.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 01:30:35 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Liu",
"Shu-Cong",
""
],
[
"Wang",
"Yong-Qiang",
""
],
[
"Zhao",
"Zhen-Hua",
""
]
] |
2512.10198
|
Yueqing Du
|
Du Yueqing, Zhang Zhenzhu, Zhang Heze, Xue Jia, Zeng Chao, Cui Yudong, Mao Dong, Malomed Boris, Zhao Jianlin
|
Heteronuclear and Homonuclear Vector Solitons in Lasers
|
To be published by Physical Review Letters
| null |
10.1103/76cq-lrb6
| null |
physics.optics nlin.PS
|
http://creativecommons.org/licenses/by/4.0/
|
Vector solitons (VSs), being observed across various fields from optics to Bose-Einstein condensates, are localized structures composed of orthogonal modes bound by nonlinear couplings. Nevertheless, the influence of intermodal linear coupling on the physical properties of this bimodal structure remains to be decently revealed and harnessed. Utilizing an ultrafast fiber laser as a platform, we predict and demonstrate that the linear mode coupling (LMC) induces the deformable VS in terms of the temporal and spectral structures. Weak LMC supports heteronuclear vector solitons built of dissimilar polarization modes, i.e., a single pulse coupled to an orthogonal damped pulse chain. On the other hand, strong LMC facilitates the homonuclear VS composed of polarization modes with similar structures, in the form of soliton compounds featuring caterpillar motions. Our findings reveal new patterns of VSs and open an effective avenue for versatile ultrafast optical sources.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 01:38:09 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Yueqing",
"Du",
""
],
[
"Zhenzhu",
"Zhang",
""
],
[
"Heze",
"Zhang",
""
],
[
"Jia",
"Xue",
""
],
[
"Chao",
"Zeng",
""
],
[
"Yudong",
"Cui",
""
],
[
"Dong",
"Mao",
""
],
[
"Boris",
"Malomed",
""
],
[
"Jianlin",
"Zhao",
""
]
] |
2512.10199
|
Samuel Johnston
|
Samuel G. G. Johnston and Rohan Shiatis
|
Explicit correlation functions for the six-vertex model in the free-fermion regime
|
27 pages
| null | null | null |
math.PR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this article, we show that, in the free-fermion regime of the six-vertex model, all $k$-point correlation functions of vertex types admit a determinantal representation: \begin{align*} \mathbb{P}\Bigg( \bigcap_{p=1}^k \{ \text{vertex at } v^p \text{ has type } t_p \} \Bigg) = \left( \prod_{p=1}^k a_{t_p} \right)
\det\big[ L(x^i,y^j) \big]_{i,j=1}^{2k}, \end{align*} where $t_1,\ldots,t_k \in \{1,\ldots,6\}$ label the six possible vertex types, and $\{a_t : 1 \leq t \leq 6\}$ are the corresponding six-vertex weights. For each $1 \leq p \leq k$, the four points $x^{2p-1}, x^{2p}, y^{2p-1}, y^{2p} \in (\mathbb{Z}/2)^2$ are $t_p$-dependent choices among the midpoints of the edges incident to $v^p$. The correlation kernel $L$ has the contour integral representation \begin{align*} L(x,y) = \oint_{|w_1|=1} \oint_{|w_2|=1}
\frac{dw_1}{2\pi i\, w_1}\,
\frac{dw_2}{2\pi i\, w_2}\,
w_1^{\,y_1 - x_1}\, w_2^{\,y_2 - x_2}\,
h\big(c(x),c(y);w_1,w_2\big), \end{align*} where $h\big(c(x),c(y);w_1,w_2\big)$ is a simple rational function of $(w_1,w_2)$ that depends on $x$ and $y$ only through their orientations $c(x)$ and $c(y)$. Our proof is fully self-contained: we construct a determinantal point process on $\mathbb{Z}^2$ and identify the six-vertex model as its pushforward under an explicit mapping.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 01:39:01 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Johnston",
"Samuel G. G.",
""
],
[
"Shiatis",
"Rohan",
""
]
] |
2512.10200
|
Alex Colling
|
Alex Colling
|
Symmetries of extremal horizons
|
29 pages
| null | null | null |
gr-qc hep-th math.DG
|
http://creativecommons.org/licenses/by/4.0/
|
We prove an intrinsic analogue of Hawking's rigidity theorem for extremal horizons in arbitrary dimensions: any compact cross-section of a rotating extremal horizon in a spacetime satisfying the null energy condition must admit a Killing vector field. If the dominant energy condition is satisfied for null vectors, it follows that an extension of the near-horizon geometry admits an enhanced isometry group containing $SO(2,1)$ or the 2D Poincar\'e group $\mathbb{R}^2 \rtimes SO(1,1)$. In the latter case, the associated Aretakis instability for a massless scalar field is shifted by one order in the derivatives of the field transverse to the horizon. We consider a broad class of examples including Einstein-Maxwell(-Chern-Simons) theory and Yang-Mills theory coupled to charged matter. In these examples we show that the symmetries are inherited by the matter fields.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 01:39:35 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Colling",
"Alex",
""
]
] |
2512.10201
|
Huan Li
|
Huan Li
|
Switching of topological phase transition from semiconductor to ideal Weyl states in Cu$_2$SnSe$_3$ family of materials
|
8 pages, 8 figures
| null | null | null |
cond-mat.mtrl-sci cond-mat.str-el
|
http://creativecommons.org/licenses/by/4.0/
|
The exploration of topological phase transition (TPT) mechanisms constitutes a central theme in quantum materials research. Conventionally, transitions between Weyl semimetals (WSMs) and other topological states rely on the breaking of time-reversal symmetry (TRS) or precise manipulation of lattice symmetry, thus constraints the available control strategies and restrict the scope of viable material systems. In this work, we propose a novel mechanism for TPT that operates without TRS breaking or lattice symmetry modification: a class of semiconductors can be directly transformed into an ideal WSM via bandgap closure. This transition is driven by chemical doping, which simultaneously modulates the band gap and enhances spin-orbit coupling (SOC), leading to band inversion between the valence and conduction bands and thereby triggering the TPT. Using first-principles calculations, we demonstrate the feasibility of this mechanism in the Cu$_2$SnSe$_3$ family of materials, where two pairs of Weyl points emerge with energies extremely close to the Fermi level. The bulk Fermi surface becomes nearly point-like, while the surface Fermi surface consists exclusively of Fermi arcs. This symmetry-independent mechanism bypasses the constraints of conventional symmetry-based engineering, and also offers an ideal platform for probing the anomalous transport properties of WSMs.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 01:45:55 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Li",
"Huan",
""
]
] |
2512.10202
|
Jun Hu
|
Jun Hu and Huansheng Li
|
Standard symmetrizing forms and Murphy bases of the cyclotomic Hecke algebras
| null | null | null | null |
math.RT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this paper we compute the explicit value of the standard symmetrizing form of the cyclotomic Hecke algebra of type $G(\ell,1,n)$ on each Murphy basis element. We do this for both the non-degenerate cyclotomic Hecke algebra of type $G(\ell,1,n)$ and the degenerate cyclotomic Hecke algebra of type $G(\ell,1,n)$. As an application, we obtain some equations on the coefficients which occur in the transition matrix between the Murphy basis elements and the seminormal basis elements of the cyclotomic Hecke algebra of type $G(\ell,1,n)$.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 01:46:45 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Hu",
"Jun",
""
],
[
"Li",
"Huansheng",
""
]
] |
2512.10203
|
Abhimanyu Nag Jr
|
Abhimanyu Nag
|
On Sybil Proofness in Competitive Combinatorial Exchanges
|
26 Pages, Accepted to FC26 as a poster
| null | null | null |
econ.TH cs.CR
|
http://creativecommons.org/licenses/by/4.0/
|
We study Sybil manipulation in BRACE, a competitive equilibrium mechanism for combinatorial exchanges, by treating identity creation as a finite perturbation of the empirical distribution of reported types. Under standard regularity assumptions on the excess demand map and smoothness of principal utilities, we obtain explicit linear bounds on price and welfare deviations induced by bounded Sybil invasion. Using these bounds, we prove a sharp contrast: strategyproofness in the large holds if and only if each principal's share of identities vanishes, whereas any principal with a persistent positive share can construct deviations yielding strictly positive limiting gains. We further show that the feasibility of BRACE fails in the event of an unbounded population of Sybils and provide a precise cost threshold that ensures disincentivization of such attacks in large markets.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 01:53:04 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Nag",
"Abhimanyu",
""
]
] |
2512.10204
|
Weimin Han
|
Weimin Han
|
Variational-hemivariational inequalities: A brief survey on mathematical theory and numerical analysis
| null | null | null | null |
math.NA cs.NA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Variational-hemivariational inequalities are an area full of interesting and challenging mathematical problems. The area can be viewed as a natural extension of that of variational inequalities. Variational-hemivariational inequalities are valuable for application problems from physical sciences and engineering that involve non-smooth and even set-valued relations, monotone or non-monotone, among physical quantities. In the recent years, there has been substantial growth of research interest in modeling, well-posedness analysis, development of numerical methods and numerical algorithms of variational-hemivariational inequalities. This survey paper is devoted to a brief account of well-posedness and numerical analysis results for variational-hemivariational inequalities. The theoretical results are presented for a family of abstract stationary variational-hemivariational inequalities and the main idea is explained for an accessible proof of existence and uniqueness. To better appreciate the distinguished feature of variational-hemivariational inequalities, for comparison, three mechanical problems are introduced leading to a variational equation, a variational inequality, and a variational-hemivariational inequality, respectively. The paper also comments on mixed variational-hemivariational inequalities, with examples from applications in fluid mechanics, and on results concerning the numerical solution of other types (nonstationary, history dependent) of variational-hemivariational inequalities.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 01:53:08 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Han",
"Weimin",
""
]
] |
2512.10205
|
Min Chen
|
Min Chen, Hong-Yan Song, Jia-Lin Chen, Peng Ye, Guo-Wei Zhang, Fang-Xiang Wang, Li Zhang, Shuang Wang, De-Yong He, Zhen-qiang Yin, Guang-Can Guo, Wei Chen and Zheng-Fu Han
|
Optical fuse based on the photorefractive effect for defending the light-injection attacks of quantum key distribution
|
10 pages, 5 figures
| null | null | null |
quant-ph physics.optics
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Light-injection attacks pose critical security threats to quantum key distribution (QKD) systems. Conventional defense methods, such as isolators, filters, and optical power monitoring, are confronted with the threats of specific attacks and the limitations in integration. To address this, we propose and experimentally demonstrate an integrated attack sensing and automatic response unit utilizing the photorefractive effect in a thin-film lithium niobate microring resonator. Our unit provides a high rejection ratio against non-resonant injected light. For resonant attacks exceeding tens of microwatts, the unit can autonomously attenuate the transmission of the quantum signal light, leading to a significant suppression of the secret key rate. This work enhances the security of QKD systems against light-injection attacks by providing a highly sensitive, broadband, and on-chip defense mechanism.
|
[
{
"version": "v1",
"created": "Thu, 11 Dec 2025 01:54:24 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chen",
"Min",
""
],
[
"Song",
"Hong-Yan",
""
],
[
"Chen",
"Jia-Lin",
""
],
[
"Ye",
"Peng",
""
],
[
"Zhang",
"Guo-Wei",
""
],
[
"Wang",
"Fang-Xiang",
""
],
[
"Zhang",
"Li",
""
],
[
"Wang",
"Shuang",
""
],
[
"He",
"De-Yong",
""
],
[
"Yin",
"Zhen-qiang",
""
],
[
"Guo",
"Guang-Can",
""
],
[
"Chen",
"Wei",
""
],
[
"Han",
"Zheng-Fu",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.