arXiv papers from September 2025

Linjin He, Xinda Qi, Dong Chen, Zhaojian Li

Soft snake robots offer remarkable flexibility and adaptability in complex environments, yet their control remains challenging due to highly nonlinear dynamics. Existing model-based and bio-inspired controllers rely on simplified assumptions that limit performance. Deep reinforcement learning (DRL) has recently emerged as a promising alternative, but online

Shucheng Zhang, Yan Shi, Bingzhang Wang, Yuang Zhang

Ensuring the safety of vulnerable road users (VRUs), such as pedestrians and cyclists, remains a critical global challenge, as conventional infrastructure-based measures often prove inadequate in dynamic urban environments. Recent advances in artificial intelligence (AI), particularly in visual perception and reasoning, open new opportunities for proactive a

Manuel Cebrian, Tomomi Kito, Raul Castro Fernandez

Large language models are proliferating, and so are the benchmarks that serve as their common yardsticks. We ask how the agglomeration patterns of these two layers compare: do they evolve in tandem or diverge? Drawing on two curated proxies for the ecosystem, the Stanford Foundation-Model Ecosystem Graph and the Evidently AI benchmark registry, we find compl

Nik Rollinson, Nikolaos Polatidis

Android malware continues to evolve through obfuscation and polymorphism, posing challenges for both signature-based defenses and machine learning models trained on limited and imbalanced datasets. Synthetic data has been proposed as a remedy for scarcity, yet the role of large language models (LLMs) in generating effective malware data for detection tasks r

Shaoyi Zheng, Canyu Zhang, Tianyi Zhou, Shengjie Wang

In-context learning (ICL) enables efficient few-shot learning in large language models (LLMs) without training, but suffers from the quadratic input complexity of transformers, limiting the maximum number of exemplars. While various efficient ICL approaches partition the context into blocks to process (e.g., ensembling, compression, cross-attention), they of

Islam I. Abdulaal, Abdelrahman W. A. Elsayed, Omar A. M. Abdelraouf

Bound states in the continuum (BICs) have emerged as a revolutionary paradigm in terahertz (THz) photonics, enabling metasurfaces with theoretically infinite quality factors (Q-factors) and unprecedented light-matter control. This review synthesizes a decade of progress in THz-BIC research, tracing the evolution from foundational symmetry-protected designs t

Daniel G. Williams

Vocal dereverberation remains a challenging task in audio processing, particularly for real-time applications where both accuracy and efficiency are crucial. Traditional deep learning approaches often struggle to suppress reverberation without degrading vocal clarity, while recent methods that jointly predict magnitude and phase have significant computationa

Renee Ge, Qianli Liao, Tomaso Poggio

Transformers have demonstrated remarkable performance in natural language processing and related domains, as they largely focus on sequential, autoregressive next-token prediction tasks. Yet, they struggle in logical reasoning, not necessarily because of a fundamental limitation of these models, but possibly due to the lack of exploration of more creative us

Josh Hunter, John McDermid, Simon Burton, Poppy Fynes

In the field of railway automation, one of the key challenges has been the development of effective computer vision systems due to the limited availability of high-quality, sequential data. Traditional datasets are restricted in scope, lacking the spatio temporal context necessary for real-time decision-making, while alternative solutions introduce issues re

Shicheng Liu, Qilong Zhai

In this paper, we present a posteriori error estimation for weak Galerkin method applied to fourth order singularly perturbed problem. The weak Galerkin discretization space and numerical scheme are first described. A fully computable residual type error estimator is then constructed. Both the reliability and efficiency of the proposed estimator are rigorous

Xuanyu Lyu, Jared Balbona, Tong Chen, Matthew C. Keller

Genetic nurture effects and assortative mating (AM) occur across many human behaviors and can bias estimates from traditional genetic models. These influences are typically studied univariately, within the same trait. However, estimation of cross-trait genetic nurture effects and cross-trait AM remains underexplored due to the absence of suitable approaches.

Tong Chen, Yinuo Zhang, Pranam Chatterjee

Designing sequences that satisfy multiple, often conflicting, objectives is a central challenge in therapeutic and biomolecular engineering. Existing generative frameworks largely operate in continuous spaces with single-objective guidance, while discrete approaches lack guarantees for multi-objective Pareto optimality. We introduce AReUReDi (Annealed Rectif

Rohit Dilip, Evan Zhang, Ayush Varshney, David Van Valen

Protein structure tokenizers enable the creation of multimodal models of protein structure, sequence, and function. Current approaches to protein structure tokenization rely on bespoke components that are invariant to spatial symmetries, but that are challenging to optimize and scale. We present Kanzi, a flow-based tokenizer for tokenization and generation o

Zilai Li, Lujia Bai

Diffusion models are the state-of-the-art generative models for high-resolution images, but sampling from pretrained models is computationally expensive, motivating interest in fast sampling. Although Free-U Net is a training-free enhancement for improving image quality, we find it ineffective under few-step ($<10$) sampling. We analyze the discrete diffusio

Akshaya Kumar, Anna Raymaker, Michael Specter

We conduct the first comprehensive security analysis of Tile, the second most popular crowd-sourced location-tracking service behind Apple's AirTags. We identify several exploitable vulnerabilities and design flaws, disproving many of the platform's claimed security and privacy guarantees: Tile's servers can persistently learn the location of all users and t

Felix Reichel

This paper introduces a contest-theoretic simplified model of triathlon as a sequential two-stage game. In Stage 1 (post-swim), participants decide whether to continue or withdraw from the contest, thereby generating an endogenous participation decision. In Stage 2 (bike-run), competition is represented as a Tullock contest in which swim drafting acts as a m

Alperen Tercan, Necmiye Ozay

Constrained Markov Decision Processes (CMDPs) are notably more complex to solve than standard MDPs due to the absence of universally optimal policies across all initial state distributions. This necessitates re-solving the CMDP whenever the initial distribution changes. In this work, we analyze how the optimal value of CMDPs varies with different initial dis

Huitao Yang, Guanting Chen

As large language models (LLMs) continue to grow in capability, there is increasing interest in incorporating them into decision-making tasks. A common pipeline for this is Decision-Pretrained Transformers (DPTs). However, existing training methods for DPTs often struggle to generalize beyond their pretraining data distribution. To explore mitigation of this

Yuanbo Hou, Zhaoyi Liu, Xin Shen, Stephen Roberts

Mosquito Species Classification (MSC) is crucial for vector surveillance and disease control. The collection of mosquito bioacoustic data is often limited by mosquito activity seasons and fieldwork. Mosquito recordings across regions, habitats, and laboratories often show non-biological variations from the recording environment, which we refer to as domain f

Yiping Ji, James Martens, Jianqiao Zheng, Ziqin Zhou

Transformers have achieved remarkable success across a wide range of applications, a feat often attributed to their scalability. Yet training them without skip (residual) connections remains notoriously difficult. While skips stabilize optimization, they also disrupt the hierarchical structure of representations, raising the long-standing question of whether

Liangliang Zhang, Nandana Mihindukulasooriya, Niharika S. D'Souza, Sola Shirai

Data products are reusable, self-contained assets designed for specific business use cases. Automating their discovery is of great industry interest, as it enables efficient data access in large data lakes and supports analytical workflows. However, no benchmark currently exists for data product discovery over hybrid table-text corpora. Existing datasets foc

Chang Han, Andrew Mcnutt

Superstition and religious belief system have historically shaped human behavior, offering powerful psychological motivations and persuasive frameworks to guide actions. Inspired by Feng Shui -- an ancient Chinese superstition -- this paper proposes a pseudo-theoretical framework that integrates superstition-like heuristics into visualization design. Rather

Samuel Li, Ian P. Roberts

Existing beamforming-based full-duplex solutions for multi-antenna wireless systems often rely on explicit estimation of the self-interference channel. The pilot overhead of such estimation, however, can be prohibitively high in millimeter-wave and massive MIMO systems, thus limiting the practicality of existing solutions, especially in fast-fading condition

Poonam Rani, Masayuki Mashiko, Keisuke Hirata, Ken-ichi Uchida

Thermal diode is a growing technology and important for active thermal flow control. Since the theoretical designing of thermal diode in 2004, various kinds of solid-state thermal diodes have been theoretically and experimentally investigated. Here, we report on the observation of thermal rectification in bulk-size superconductor-normal metal junctions. High

Marguerite H. Mahr, Noah D. Marshall, Jessa Marley, Sarah K. Wyse

Road mortality may be a significant factor in the global decline of amphibian populations, yet rigorous assessments of its effect on long-term population persistence are lacking. Here, we investigate population persistence through a field study and mathematical model of a western toad ({\textit{Anaxyrus Boreas}} {\RR(Baird and Girard, 1852)}) population with

T. James Brandt

Adaptive chatbots that mimic a user's linguistic style can build rapport and engagement, yet unconstrained mimicry risks an agent that feels unstable or sycophantic. We present a computational evaluation framework that makes the core design tension explicit: balancing moment-to-moment linguistic synchrony against long-term persona stability. Using an 8-dimen

Stefan Gobej

Given a fixed graph H, we say that a graph G is H-free if G does not contain H as a subgraph. The Tur\'an number ex(n, H) of H is the maximum number of edges in an n-vertex H-free graph. The study of Tur\'an number of graphs is a central topic in extremal graph theory. The purpose of this article is to present some well-known results about this field but als

Jae Woo Lee

We present high-precision TESS photometry of V421 Peg (TIC 301747091), an early F-type eclipsing binary containing a candidate $\gamma$ Dor component. The observed short-cadence data allow the detection of pulsation signals, along with revision of the fundamental properties of the component stars. Detailed binary modeling indicated that the program target is

Anirudh Subramanyam, Yuxin Chen, Robert L. Grossman

Scaling laws for language model training traditionally characterize how performance scales with model size and dataset volume. Prior work has explored architecture variants and data treatments such as dataset filtering and noise injection in language model pretraining; however, these studies have not formalized data quality within a principled scaling law. W

Lance Edward Miller, Jackson S. Morrow

In this work, we prove a quantitative version of the prime-to-$p$ Manin--Mumford conjecture for varieties with ample cotangent bundle. More precisely, let $A$ be an abelian variety defined over a number field $F$, and let $X$ be a smooth projective subvariety of $A$ with ample cotangent bundle. We prove that for every prime $p\gg 0$, the intersection of $X(F

Matthew Marshall, Jacob Williamson, Seth Hyra, Robert H. Leonard

We present a simple method for removing polyimide coatings from optical fibers using inexpensive and readily available solvents. Impacts of solvent mixing ratios, soak temperature, material expansion, wicking, and drying are described to provide empirical context for the method. We find that soaking fibers for six hours in a 2:1 mixture of methanol to aceton

Kieran Drury, Martine J. Barons, Jim Q. Smith

Many Bayesian network modelling applications suffer from the issue of data scarcity. Hence the use of expert judgement often becomes necessary to determine the parameters of the conditional probability tables (CPTs) throughout the network. There are usually a prohibitively large number of these parameters to determine, even when complementing any available d

Ayushi Dube, Gian Singh, Sarma Vrudhula

Transprecision computing (TC) is a promising approach for energy-efficient machine learning (ML) computation on resource-constrained platforms. This work presents a novel ASIC design of a Transprecision Arithmetic and Logic Unit (TALU) that can support multiple number formats: Posit, Floating Point (FP), and Integer (INT) data with variable bitwidth of 8, 16

Zeshi Dai, Zimo Peng, Zerui Cheng, Ryan Yihe Li

We present CAIA, a benchmark exposing a critical blind spot in AI evaluation: the inability of state-of-the-art models to operate in adversarial, high-stakes environments where misinformation is weaponized and errors are irreversible. While existing benchmarks measure task completion in controlled settings, real-world deployment demands resilience against ac

Panos Giannopoulos, Miriam Goetze, Grzegorz Gutowski, Maarten Löffler

Drawing graphs with the minimum number of crossings is a classical problem that has been studied extensively. Many restricted versions of the problem have been considered. For example, bipartite graphs can be drawn such that the two sets in the bipartition of the vertex set are mapped to two parallel lines, and the edges are drawn as straight-line segments.

Z. Wang, A. C. Seth, M. Latour, J. Strader

We investigate the foreground interstellar medium along the line of sight and intracluster medium of $\omega$ Centauri ($\omega$ Cen) by measuring the equivalent width of Na I D absorptions from MUSE observations. The large line-of-sight velocity difference between $\omega$ Cen and the foreground enables us to separate Na I D absorption contributed from atom

Sarmad Mehrdad, Maxime Sabbah, Vincent Bonnet, Ludovic Righetti

This paper investigates the application of Minimal Observation Inverse Reinforcement Learning (MO-IRL) to model and predict human arm-reaching movements with time-varying cost weights. Using a planar two-link biomechanical model and high-resolution motion-capture data from subjects performing a pointing task, we segment each trajectory into multiple phases a

Ahmed Fawzy, Amjed Tahir, Kelly Blincoe

AI code generation tools are transforming software development, especially for novice and non-software developers, by enabling them to write code and build applications faster and with little to no human intervention. Vibe coding is the practice where users rely on AI code generation tools through intuition and trial-and-error without necessarily understandi

Alireza Salemi, Mihir Parmar, Palash Goyal, Yiwen Song

Advances in large language models (LLMs) have created new opportunities in data science, but their deployment is often limited by the challenge of finding relevant data in large data lakes. Existing methods struggle with this: both single- and multi-agent systems are quickly overwhelmed by large, heterogeneous files, and master-slave multi-agent systems rely

Yuhui Liu, Samannita Halder, Shian Wang, Tianyi Li

This study introduces a novel control framework for adaptive cruise control (ACC) in automated driving, leveraging Long Short-Term Memory (LSTM) networks and physics-informed constraints. As automated vehicles (AVs) adopt advanced features like ACC, transportation systems are becoming increasingly intelligent and efficient. However, existing AV control strat

Mark Skandera

Given a weakly decreasing positive integer sequence $\lambda = (\lambda_1,\dotsc,\lambda_\ell)$ summing to $n$, let $\chi^\lambda$ denote the irreducible character of the symmetric group $S_n$ indexed by $\lambda$. This representation has dimension $\chi^\lambda(e)$, where $e$ is the identity element of $S_n$. Let $\mathrm{Imm}_{\chi^\lambda}$ denote the cor

Hassen Dhrif

The emergence of large language models has enabled sophisticated multi-agent systems, yet coordinating their reasoning capabilities through prompt engineering remains challenging. We present a theoretically-grounded framework for dynamic prompt orchestration that enhances reasoning across multiple specialized agents. This framework addresses three core chall

Priyank Dubey

Link prediction is one of the fundamental problems in graph theory, critical for understanding and forecasting the evolution of complex systems like social and biological networks. While classical heuristics capture certain aspects of graph topology, they often struggle to optimally integrate local and global structural information or adapt to complex depend

Lucas Roberts, Denisa Roberts

Code search is an important information retrieval application. Benefits of better code search include faster new developer on-boarding, reduced software maintenance, and ease of understanding for large repositories. Despite improvements in search algorithms and search benchmarks, the domain of code search has lagged behind. One reason is the high cost of hum

Tamara Caldas Cifuentes, Jemma Harris, Patricia Verheyden

CERN's new flagship education and outreach centre, Science Gateway, opened its doors to the public in autumn 2023. Through a combination of immersive scenography with interactive exhibits and real scientific objects, its permanent exhibitions address the organisation's particle physics research and how this knowledge applies to other scientific fields and th

Katarzyna W. Kowalik

We show that the theory $\mathsf{WKL}^*_0+\mathsf{CAC}$ is polynomially simulated by $\mathsf{RCA}_0^*$ with respect to $\forall\Pi^0_3$ formulas. For the proof, we use the method of forcing interpretations and syntactically simulate a two-step model-theoretic argument, which involves construction of a restricted definable ultrapower, followed by a generic c

Günter F. Steinke, Thomas Steinke

Standard techniques for differentially private estimation, such as Laplace or Gaussian noise addition, require guaranteed bounds on the sensitivity of the estimator in question. But such sensitivity bounds are often large or simply unknown. Thus we seek differentially private methods that can be applied to arbitrary black-box functions. A handful of such tec

A. K. Hamisu, K. Jasleen

The exponential growth of internet generated data has fueled advancements in artificial intelligence (AI), machine learning (ML), and deep learning (DL) for extracting actionable insights in marketing,telecom, and health sectors. This chapter explores ML applications across three domains namely healthcare, marketing, and telecommunications, with a primary fo

Haoran Xi, Minghao Shao, Brendan Dolan-Gavitt, Muhammad Shafique

Large language models show promise for vulnerability discovery, yet prevailing methods inspect code in isolation, struggle with long contexts, and focus on coarse function- or file-level detections that offer limited guidance to engineers who need precise line-level localization for targeted patches. We introduce T2L, an executable framework for project-leve

Yuhui Liu, Shian Wang, Ansel Panicker, Kate Embry

Internal combustion engine (ICE) vehicles and electric vehicles (EVs) exhibit distinct vehicle dynamics. EVs provide rapid acceleration, with electric motors producing peak power across a wider speed range, and achieve swift deceleration through regenerative braking. While existing microscopic models effectively capture the driving behavior of ICE vehicles,

K. Albert, J. Hirzberger, N. A. Krivova, X. Li

Small-scale magnetic flux concentrations contribute significantly to the brightness variations of the Sun, yet observing them - particularly their magnetic field - near the solar limb remains challenging. Solar Orbiter offers an unprecedented second vantage point for observing the Sun. When combined with observations from the perspective of Earth, this enabl

Vinh Nguyen

A fundamental theorem of linear programming states that a feasible linear program is solvable if and only if its objective function is copositive with respect to the recession cone of its feasible set. This paper demonstrates that this crucial guarantee does not extend to Second-Order Cone Programs (SOCPs), a workhorse model in robust and convex optimization

Youpeng Li, Kartik Joshi, Xinda Wang, Eric Wong

The widespread adoption of open-source software (OSS) necessitates the mitigation of vulnerability risks. Most vulnerability detection (VD) methods are limited by inadequate contextual understanding, restrictive single-round interactions, and coarse-grained evaluations, resulting in undesired model performance and biased evaluation results. To address these

Mohammad Rostami, Atik Faysal, Reihaneh Gh. Roshan, Huaxia Wang

Identifying wireless modulation schemes is essential for cognitive radio, but standard supervised models often degrade under distribution shift, and training domain-specific wireless foundation models from scratch is computationally prohibitive. Large Language Models (LLMs) offer a promising training-free alternative via in-context learning, yet feeding raw

Haoyue Bai, Haoyu Wang, Shengyu Chen, Zhengzhang Chen

Large Language Models (LLMs) have shown remarkable performance on general Question Answering (QA), yet they often struggle in domain-specific scenarios where accurate and up-to-date information is required. Retrieval-Augmented Generation (RAG) addresses this limitation by enriching LLMs with external knowledge, but existing systems primarily rely on unstruct

Michael R. Powers

From a well-known equation of Hardy, one can derive a simple linear combination of the Euler-Mascheroni constant ($\gamma=0.577215\ldots$) and Euler-Gompertz constant ($\delta=0.596347\ldots$): $\gamma+\delta/e=\textrm{Ein}\left(1\right)$. Although neither $\gamma$ nor $\delta$ is currently known to be irrational, this linear combination has been shown to be

Xiaotang Zhang, Ziyi Chang, Qianhui Men, Hubert P. H. Shum

Motion in-betweening is the problem to synthesize movement between keyposes. Traditional research focused primarily on single characters. Extending them to densely interacting characters is highly challenging, as it demands precise spatial-temporal correspondence between the characters to maintain the interaction, while creating natural transitions towards p

Tanmay Khandelwal, Magdalena Fuentes

Diffusion Transformers (DiTs) enable high-quality audio synthesis but are often computationally intensive and require substantial storage, which limits their practical deployment. In this paper, we present a comprehensive evaluation of post-training quantization (PTQ) techniques for audio DiTs, analyzing the trade-offs between static and dynamic quantization

Matthew David Hamilton

Artificial intelligence is set to revolutionize social and political life in unpredictable ways, raising questions about the principles that ought to guide its development and regulation. By examining digital advertising and social media algorithms, this article highlights how artificial intelligence already poses a significant threat to the republican conce

Bowen Wei, Yuan Shen Tay, Howard Liu, Jinhao Pan

Security Operations Centers (SOCs) are overwhelmed by tens of thousands of daily alerts, with only a small fraction corresponding to genuine attacks. This overload creates alert fatigue, leading to overlooked threats and analyst burnout. Classical detection pipelines are brittle and context-poor, while recent LLM-based approaches typically rely on a single m

Akash Dhasade, Sadegh Farhadkhani, Rachid Guerraoui, Nirupam Gupta

Federated inference, in the form of one-shot federated learning, edge ensembles, or federated ensembles, has emerged as an attractive solution to combine predictions from multiple models. This paradigm enables each model to remain local and proprietary while a central server queries them and aggregates predictions. Yet, the robustness of federated inference

Zhongxuan Liu, Yue Kang, Thomas C. M. Lee

The Lipschitz bandit problem extends stochastic bandits to a continuous action set defined over a metric space, where the expected reward function satisfies a Lipschitz condition. In this work, we introduce a new problem of Lipschitz bandit in the presence of stochastic delayed feedback, where the rewards are not observed immediately but after a random delay

Panagiotis Kounatidis, Andreas A. Malikopoulos

In this paper, we present the combined learning-and-control (CLC) approach, which is a new way to solve optimal control problems with unknown dynamics by unifying model-based control and data-driven learning. The key idea is simple: we design a controller to be optimal for a proxy objective built on an available model while penalizing mismatches with the rea

Rong Liu, Zhongpai Gao, Benjamin Planche, Meida Chen

We introduce Universal Beta Splatting (UBS), a unified framework that generalizes 3D Gaussian Splatting to N-dimensional anisotropic Beta kernels for explicit radiance field rendering. Unlike fixed Gaussian primitives, Beta kernels enable controllable dependency modeling across spatial, angular, and temporal dimensions within a single representation. Our uni

Thierry Blankenstein, Jialin Yu, Zixuan Li, Vassilis Plachouras

Agents backed by large language models (LLMs) increasingly rely on external tools drawn from marketplaces where multiple providers offer functionally equivalent options. This raises a critical fairness concern: systematic bias in tool selection can degrade user experience and distort competition by privileging certain providers over others. We introduce a be

Wenyang Jia, Jingjing Wang, Kai Lei

Modern blockchains need fast, reliable propagation to balance security and throughput. Virtual-coordinate methods speed dissemination but rely on slow iterative updates, leaving nodes out of sync. We present BlockSDN-VC, a transaction-broadcast protocol that centralises coordinate computation and forwarding control in an SDN controller, delivering global con

Raphael Rousset-Zenou, Nicolas Aparicio, Simon Messelot, Rasmus D. Schlosser

Superconductor-semiconductor hybrid materials have been extensively used for experiments on electrically tunable quantum devices. Notably, Josephson junctions utilizing nanowire weak links have enabled a number of new gate-tunable qubits, including gatemons, Andreev level qubits and spin qubits. Conversely, superconducting parametric amplifiers based on Jose

Anay Majee, Amitesh Gangrade, Rishabh Iyer

Open-World Object Detection (OWOD) enriches traditional object detectors by enabling continual discovery and integration of unknown objects via human guidance. However, existing OWOD approaches frequently suffer from semantic confusion between known and unknown classes, alongside catastrophic forgetting, leading to diminished unknown recall and degraded know

FAIR CodeGen team, Jade Copet, Quentin Carbonneaux, Gal Cohen

We release Code World Model (CWM), a 32-billion-parameter open-weights LLM, to advance research on code generation with world models. To improve code understanding beyond what can be learned from training on static code alone, we mid-train CWM on a large amount of observation-action trajectories from Python interpreter and agentic Docker environments, and pe

Mohammed Alghadeer, Khanh Uyen Giang, Shuxiang Cao, Simone D. Fasciati

Algorithmic cooling shows that it is possible to locally reduce the entropy of a qubit belonging to an isolated ensemble such as nuclear spins in molecules or nitrogen-vacancy centers in diamonds. In the same physical setting, we introduce double-bracket algorithmic cooling (DBAC), a protocol that systematically suppresses quantum coherence of pure states. D

Qingshu Xu, Hong Jiao, Tianyi Zhou, Ming Li

Accurate alignment of items to content standards is critical for valid score interpretation in large-scale assessments. This study evaluates three automated paradigms for aligning items with four domain and nineteen skill labels. First, we extracted embeddings and trained multiple classical supervised machine learning models, and further investigated the imp

David J. Hemmer, Armin Straub, Karlee J. Westrem

We prove a series of ``knapsack'' type equalities for irreducible character degrees of symmetric groups. That is, we find disjoint subsets of the partitions of $n$ so that the two corresponding character-degree sums are equal. Our main result refines our recent description of the Riordan numbers as the sum of all character degrees $f^\lambda$ where $\lambda$

Serena Gomez Wannaz

ICL guides are known to improve task-specific performance, but their impact on cross-domain cognitive abilities remains unexplored. This study examines how ICL guides affect reasoning across different knowledge domains using six variants of the GPT-OSS:20b model: one baseline model and five ICL configurations (simple, chain-of-thought, random, appended text,

Han Wang, Haoyu Li, Brian Ko, Huan Zhang

Leaderboards for LRMs have turned evaluation into a competition, incentivizing developers to optimize directly on benchmark suites. A shortcut to achieving higher rankings is to incorporate evaluation benchmarks into the training data, thereby yielding inflated performance, known as benchmark contamination. Surprisingly, our studies find that evading contami

J. Sanz-Forcada, E. González-Álvarez, M. R. Zapatero Osorio, J. A. Caballero

Aims. We aim to confirm and measure the mass of the transiting planet candidate around the K5V star TOI-2093, previously announced by the Transiting Exoplanet Survey Satellite (TESS) project. Methods. We combined photometric data from 32 sectors between 2019 and 2024 with 86 radial velocity measurements obtained with the CARMENES spectrograph over a period o

Changjie Lu, Sourya Sengupta, Hua Li, Mark A. Anastasio

Objective, task-based measures of image quality (IQ) have been widely advocated for assessing and optimizing medical imaging technologies. Besides signal detection theory-based measures, information-theoretic quantities have been proposed to quantify task-based IQ. For example, task-specific information (TSI), defined as the mutual information between an ima

Vikram Krishnamurthy, Luke Snow

We study counterfactual stochastic optimization of conditional loss functionals under misspecified and noisy gradient information. The difficulty is that when the conditioning event has vanishing or zero probability, naive Monte Carlo estimators are prohibitively inefficient; kernel smoothing, though common, suffers from slow convergence. We propose a two-st

Guy Bar-Shalom, Fabrizio Frasca, Yaniv Galron, Yftah Ziser

Detecting hallucinations in Large Language Model-generated text is crucial for their safe deployment. While probing classifiers show promise, they operate on isolated layer-token pairs and are LLM-specific, limiting their effectiveness and hindering cross-LLM applications. In this paper, we introduce a novel approach to address these shortcomings. We build o

Charlotte Li, Nick Hagar, Sachita Nishal, Jeremy Gilbert

Benchmarks play a significant role in how technology companies communicate about model capabilities and how researchers and the public understand generative AI systems. However, existing benchmarks have been criticized for their failure to adequately capture real-world usages (i.e. ecological validity) or to measure underlying concepts (i.e. construct validi

Bishnu Paudel, Kathleen Petersen, Haiyang Wang

We explore the dependence of the minimal integral Mahler measure of Galois quartic fields on the discriminant of the field. We obtain density results which are conditional on the ABC conjecture as well as several unconditional results.

Shutong Wu, Jiawei Zhang

Diffusion Large Language Models (DLLMs) have emerged as a new paradigm of language modeling beyond autoregressive next-token prediction. Taking advantage of their inherent modeling foundations, DLLMs have the great potential of efficient inference with parallel decoding algorithms, which enable multi-token prediction. However, the high generation quality oft

Samar Fares, Nurbek Tastan, Noor Hussein, Karthik Nandakumar

Generative models can generate photorealistic images at scale. This raises urgent concerns about the ability to detect synthetically generated images and attribute these images to specific sources. While watermarking has emerged as a possible solution, existing methods remain fragile to realistic distortions, susceptible to adaptive removal, and expensive to

John Carney, Everett Meike

Generalizing unknotting number, $n$-adjacent knots have $n$ crossings such that changing any non-empty subset of them results in the unknot. In this paper, we determine the 2-adjacent knots through 12 crossings. Using Heegaard Floer $d$-invariants and the Alexander polynomial, we develop a new technique to obstruct 2-adjacency, and we prove conjectures of It

Hsueh-Yung Lin, Evgeny Shinder

We introduce horizontal and vertical motivic invariants of birational maps between rational dominant maps and study their basic properties. As a first application, we show that the (usual) motivic invariants vanish for birational automorphisms of threefolds over algebraically closed fields of characteristic zero. On the other hand, we prove that the motivic

Francesco Costa, Jaime Hoefken Zink, Michele Lucente, Silvia Pascoli

We present ELENA (EvaLuator of tunnElliNg Actions), an open-source Python package designed to compute the full evolution of first-order phase transitions in the early Universe generated by particle physics models, taking into account several refinements that go beyond commonly assumed simplifications. The core of ELENA is based on a vectorized implementation

Joachim P. Leibold, Lina M. Todenhagen, Matthias Althammer, Nikhita Khera

Nitrogen-vacancy (NV) centers in diamond are optically addressable spin defects with great potential for nanoscale quantum sensing. A key application of NV centers is the detection of external spins at the diamond surface. Among metals, platinum thin films - widely used in spintronics, catalysis and electrochemistry - provide a particularly interesting syste

Ľuboš Kriš, Jaroslav Kopčan, Qiwei Peng, Andrej Ridzik

The proliferation of transformer-based language models has revolutionized NLP domain while simultaneously introduced significant challenges regarding model transparency and trustworthiness. The complexity of achieving explainable systems in this domain is evidenced by the extensive array of explanation methods and evaluation metrics developed by researchers.

Wei Li, Nilanjan Chakraborty, Robert Lunde

We consider statistical inference for network-linked regression problems, where covariates may include network summary statistics computed for each node. In settings involving network data, it is often natural to posit that latent variables govern connection probabilities in the graph. Since the presence of these latent features makes classical regression as

Anna Lenzhen

We study limit sets of Teichm\"uller disks in the Thurston boundary of Teichm\"uller space of a closed surface S of genus at least 2. It is well known that almost every Teichm\"uller geodesic ray converges to a point on the boundary. We show that unlike rays, Teichm\"uller disks with smallest possible limit sets are extremely rare.

Joseph Kapusta, Mayank Singh, Thomas Welle

We introduce the equations of relativistic hydrodynamics that incorporate phase separation via spinodal decomposition. These equations consider surface effects between the two phases and are applicable for simulating intermediate-energy heavy-ion collisions and binary neutron star mergers, where a first-order phase transition is expected. We solve these equa

Illia Lukin, Andrii Sotnikov

We study SU(4)-symmetric Heisenberg model on the cubic lattice with spatially anisotropic magnetic couplings. We utilize several approaches based on the tensor-network representation of the many-body wave functions, which enable accurate analysis of ground-state properties of the model in different regimes of spatial anisotropy including fully isotropic thre

Chetwin Low, Weimin Wang, Calder Katyal

Audio-video generation has often relied on complex multi-stage architectures or sequential synthesis of sound and visuals. We introduce Ovi, a unified paradigm for audio-video generation that models the two modalities as a single generative process. By using blockwise cross-modal fusion of twin-DiT modules, Ovi achieves natural synchronization and removes th

Huanshu Zhang, Lei Kang, Sawyer D. Campbell, Jacob T. Young

Data-driven approaches have revolutionized the design and optimization of photonic metadevices by harnessing advanced artificial intelligence methodologies. This review takes a model-centric perspective that synthesizes emerging design strategies and delineates how traditional trial-and-error and computationally intensive electromagnetic simulations are bein

George Miloshevich, Luka Vranckx, Felipe Nathan de Oliveira Lopes, Pietro Dazzi

In this work, we introduce a non-local five-moment electron pressure tensor closure parametrized by a Fully Convolutional Neural Network (FCNN). Electron pressure plays an important role in generalized Ohm's law, competing with electron inertia. This model is used in the development of a surrogate model for a fully kinetic energy-conserving semi-implicit Par

Lie Qian

We describe the set of points of the trianguline variety over a given local Galois representation. Global analogues describing companion points in eigenvariety by [Bre14] and [HN17], can be thought of as a rational analogue to the weight part of Serre's conjecture. Along the same line, local companion points conjecture can be thought of as a rational analogu

Ruochen Li, Jun Li, Bailiang Jian, Kun Yuan

Automatically generated radiology reports often receive high scores from existing evaluation metrics but fail to earn clinicians' trust. This gap reveals fundamental flaws in how current metrics assess the quality of generated reports. We rethink the design and evaluation of these metrics and propose a clinically grounded Meta-Evaluation framework. We define

Israel Abebe Azime, Tadesse Destaw Belay, Atnafu Lambebo Tonja

Large Language Models (LLMs) powered with argentic capabilities are able to do knowledge-intensive tasks without human involvement. A prime example of this tool is Deep research with the capability to browse the web, extract information and generate multi-page reports. In this work, we introduce an evaluation sheet that can be used for assessing the capabili

Trung Hoang Le, Tran Cao Son, Huiping Cao

Logical rule-based methods offer an interpretable approach to knowledge graph completion (KGC) by capturing compositional relationships in the form of human-readable inference rules. While existing logical rule-based methods learn rule confidence scores, they typically assign a global weight to each rule schema, applied uniformly across the graph. This is a

Harshan Reddy Gopidi, Alae Eddine Lakraychi, Abhishek A. Panchal, Yiming Chen

The position of mobile active and inactive ions, specifically ion insertion sites, within organic crystals, significantly affects the properties of organic materials used for energy storage and ionic transport. Identifying the positions of these atom (and ion) sites in an organic crystal is difficult, especially when the element has a low X-ray scattering po

M. Huerta-Sandoval, K. Uriostegui, I. Ramos-Prieto, F. Soto-Eguibar

We present a framework for the paraxial wave equation based on propagation-dependent unitary transformations closely related to the Lewis-Ermakov invariant. This approach establishes a formal equivalence between free-space propagation and the dynamics in a quadratic gradient index (GRIN) medium. In this context, the dynamical invariant and the free-space Ham