April 2026 arXiv papers — page 154

Wenkai Li, Fan Yang, Shaunak A. Mehta, Koichi Onoue

Personality imbuing customizes LLM behavior, but safety evaluations almost always study prompt-based personas alone. We show this is incomplete: prompting and activation steering expose *different*, architecture-dependent vulnerability profiles, and testing with only one method can miss a model's dominant failure mode. Across 5,568 judged conditions on four

Tiantian Zhang, Jierui Zuo, Michael Chen, Wenping Wang

Recent theory suggests that reward-model-first methods can be more sample-efficient than direct policy fitting when the reward function is statistically simpler than the induced policy. We propose DDO-RM, a finite-candidate decision-optimization method that converts reward scores into an explicit target distribution. Unlike PPO-based RLHF or DPO, DDO-RM perf

Vikrant Malik, Taylan Kargin, Babak Hassibi

K-means clustering is a workhorse of unsupervised learning, but it is notoriously brittle to outliers, distribution shifts, and limited sample sizes. Viewing k-means as Lloyd--Max quantization of the empirical distribution, we develop a distributionally robust variant that protects against such pathologies. We posit that the unknown population distribution l

Diptasikha Das, A. Jana, S. Mahakal, Pallabi Sardar

The solid state reaction method is employed to synthesize Sn1-xInxTe samples. Power Factors of synthesized samples are estimated from resistivity and thermopower data. Modifications in structural parameters, resistivity and thermopower owing to In doping in SnTe thermoelectric material are reported. In-depth structural analysis, employing Rietveld refinement

Satyajit Puhan, Shubham Sharma, Narinder Kumar, Harleen Dahiya

Based on a global fit to experimental measurements of the pion electromagnetic form factor and parton distribution functions (PDFs), we report a data-driven determination of the unpolarized quark generalized parton distributions (GPDs) for the case of pion in the zero-skewness limit ($\xi = 0$). The form factor is parameterized using a flexible functional fo

Jianbo Cui, Mihály Kovács, Derui Sheng

We study the numerical approximation of a class of degenerate parabolic stochastic partial differential equations on non-compact metric graphs, which naturally arise in the asymptotic analysis of Hamiltonian flows under small noise perturbations. The numerical discretization of these equations faces several challenges, including the non-compactness of the gr

Kei Funano

We establish two universal inequalities for Dirichlet eigenvalues of the Laplacian on a Euclidean convex domain.

Luca Cappelli, Sauro Succi, Monica Lacatus, Alessandro Zecchi

The search for quantum-like wave formulations of the Navier-Stokes (Schr\"odinger-Navier-Stokes, SNS for short) equations describing classical dissipative fluids has met with increasing attention in the recent years, due to the large portfolio of potential applications in science and engineering. A SNS formulation of classical fluids was first presented in a

Linjie Li, Huiyu Xiao, Jiarui Cao, Zhenyu Wu

Class-incremental learning (CIL) aims to continuously accumulate knowledge from a stream of tasks and construct a unified classifier over all seen classes. Although pretrained models (PTMs) have shown promising performance in CIL, they still struggle with the entanglement of multi-task subspaces, leading to catastrophic forgetting when task routing parameter

Hatem M. El-boghdadi, Toqeer Ali Syed, Ali Akarma, Qamar Wali

The rapid adoption of AI tools such as ChatGPT has significantly transformed academic practices, offering considerable benefits for both students and faculty in computing disciplines. These tools have been shown to enhance learning efficiency, academic self-efficacy, and confidence. However, their increasing use also raises pressing concerns regarding the pr

Jialing Wang, Yue Zhao, Yuhao Zhang, Jing Yu

Recent advances in Speech Large Language Models (Speech-LLMs) have made significant progress, greatly enhancing multimodal interaction capabilities.However, their application in low-resource and dialect-diverse environments still faces challenges. The severe scarcity of Tibetan data, coupled with the phonetic differences among its major dialects (\"U-Tsang,

Daniel Nichols, Konstantinos Parasyris, Caetano Melone, Tal Ben-Nun

As high-performance computing and AI workloads become increasingly dependent on GPUs, maintaining high performance across rapidly evolving hardware generations has become a major challenge. Developers often spend months tuning scientific applications to fully exploit new architectures, navigating a complex optimization space that spans algorithm design, sour

Sidhanta Mohanty, Shaunak Sen

An important design principle for biological oscillators divides the oscillators into two classes: fixed frequency, variable amplitude and fixed amplitude, variable frequency. Because of the interplay of nonlinearity and feedback, both positive and negative, analytical investigations of this design principle are primarily based on numerical simulations of or

Xinyu Li, Yintong Huo, Chenxi Mao, Shiwen Shan

Log-based anomaly detection is fundamentally constrained by training data sparsity. Our empirical study reveals that public benchmark datasets cover less than 10% of source code log templates. Consequently, models frequently misclassify unseen but valid execution paths as anomalies, leading to false alarms. To address this, we propose AnomalyGen, a novel fra

Yuzheng Xu, Annya Dahmani, Matthew D. Blanchard, Niclas Dern

Human-AI complementarity, the idea that combining human and AI judgments can outperform either alone, offers a promising pathway toward robust oversight of advanced AI systems. However, whether human-AI complementarity can be achieved on realistic tasks remains an open question. We investigate this through two approaches: hybridization and two AI assistance

Jaewook Lee, Ernest K. Ryu, Chulhee Yun

We propose Nesterov acceleration with Operator Decomposition (NOD), which extends Nesterov's accelerated gradient descent (NAG) from smooth strongly convex optimization to the broader setting of strongly monotone, Lipschitz operators. The key insight is to decompose the operator into cyclically monotone and monotone components, with the Asplund decomposition

Pierre Jourlin

This paper presents an empirical study of a multi-model zero-shot pipeline for knowledge graph construction and exploitation, executed entirely through local inference on consumer-grade hardware. We propose a reproducible evaluation framework integrating two external benchmarks (DocRED, HotpotQA), WebQuestionsSP-style synthetic data, and the RAGAS evaluation

Xi Chen, Wei Xue, Yike Guo

Role-playing has garnered rising attention as it provides a strong foundation for human-machine interaction and facilitates sociological research. However, current work is confined to textual modalities, neglecting speech, which plays a predominant role in daily life, thus limiting genuine role-playing. To bridge this gap, we conceptualize and benchmark spee

Junfu Pu, Yuxin Chen, Teng Wang, Ying Shan

Current multimodal large language models (MLLMs) have demonstrated remarkable capabilities in short-form video understanding, yet translating long-form cinematic videos into detailed, temporally grounded scripts remains a significant challenge. This paper introduces the novel video-to-script (V2S) task, aiming to generate hierarchical, scene-by-scene scripts

Geordie Williamson, Oded Yacobi, Paul Zinn-Justin

We present a new method for constructing Hadamard matrices that combines transformer neural networks with local search in the PatternBoost framework. Our approach is designed for extremely sparse combinatorial search problems and is particularly effective for Hadamard matrices of Goethals--Seidel type, where Fourier methods permit fast scoring and optimisati

Huisheng Wang, H. Vicky Zhao

Herding, where investors imitate others' decisions rather than relying on their own analysis, is a prevalent phenomenon in financial markets. Excessive herding distorts rational decisions, amplifies volatility, and can be exploited by manipulators to harm the market. Traditional regulatory tools, such as information disclosure and transaction restrictions, a

Anagha Gopinath, Faiha Mujeeb, Subhabrata Dhar, Jyoti Mohanty

Stacking order in bilayers of transition metal dichalcogenides (TMDs) controls structural symmetry and layer-to-layer interactions, offering a direct route to tune their electronic properties and enable optoelectronic applications. The work function is a key parameter that determines the electronic and optoelectronic device performance. However, a comprehens

Zeyi Ren, Jialin Dong, Wei Zuo, Yikun Wang

Large-scale three-dimensional (3D) scene reconstruction in low-altitude intelligent networks (LAIN) demands highly efficient wireless image transmission. However, existing schemes struggle to balance severe pilot overhead with the transmission accuracy required to maintain reconstruction fidelity. To strike a balance between efficiency and reliability, this

Rongjia Yu, Tong Jia, Hao Wang, Xiaofang Li

Monocular depth estimation is a fundamental yet challenging task in computer vision, especially under complex conditions such as textureless surfaces, transparency, and specular reflections. Recent diffusion-based approaches have significantly advanced performance by reformulating depth prediction as a denoising process in the latent space. However, existing

Yan Zhou, Qingkai Fang, Yun Hong, Yang Feng

Currently, large language models (LLMs) predominantly focus on the text modality. To enable more natural human-AI interaction, speech LLMs are emerging, but building effective end-to-end speech LLMs remains challenging due to limited data and the difficulty in expanding to more languages. In this paper, we introduce Cross-lingual Speech Language Model (CSLM)

Siyu Sun, Jing Ren, Zhaohe Liao, Dongxiao Mao

Adapting decoder-only multimodal large language models (MLLMs) for unified multimodal retrieval faces two structural gaps. First, existing methods rely on implicit pooling, which overloads the hidden state of a standard vocabulary token (e.g., <EOS>) as the sequence-level representation, a mechanism never designed for information aggregation. Second, contras

Lingzhe Zhang, Yunpeng Zhai, Tong Jia, Minghua He

Contemporary microservice systems continue to grow in scale and complexity, leading to increasingly frequent and costly failures. While recent LLM-based auto-remediation approaches have emerged, they primarily translate textual instructions into executable Ansible playbooks and rely on expert-crafted prompts, lacking runtime knowledge guidance and depending

Sergio Gómez, David Hewett, Andrea Moiola

We formulate, analyse, and implement a discontinuous Galerkin finite element method (DG-FEM) for the approximation of the solution of an elliptic boundary value problem in a domain with fractal boundary. We consider the case of the Poisson equation in the Koch snowflake domain with zero Dirichlet boundary conditions, but our methodology can be generalised to

Hyewon Choi, Jooyoung Choi, Hansol Jang, Hyun Kim

Neural retrievers are often trained on large-scale triplet data comprising a query, a positive passage, and a set of hard negatives. In practice, hard-negative mining can introduce false negatives and other ambiguous negatives, including passages that are relevant or contain partial answers to the query. Such label noise yields inconsistent supervision and c

Linjie Li, Zhenyu Wu, Huiyu Xiao, Yang Ji

Prompt-based class-incremental learning methods typically construct a prompt pool consisting of multiple trainable key-prompts and perform instance-level matching to select the most suitable prompt embeddings, which has shown promising results. However, existing approaches face several limitations, including fixed prompt pools, manual selection of prompt emb

Jeremy Dao, Alan Fern

Simulation trained legged locomotion policies often exhibit performance loss on hardware due to dynamics discrepancies between the simulator and the real world, highlighting the need for approaches that adapt the simulator itself to better match hardware behavior. Prior work typically quantify these discrepancies through precise, time-aligned matching of joi

Linggang Kong, Lei Wu, Yunlong Zhang, Xiaofeng Zhong

Despite the groundbreaking advancements made by large language models (LLMs), hallucination remains a critical bottleneck for their deployment in high-stakes domains. Existing classification-based methods mainly rely on static and passive signals from internal states, which often captures the noise and spurious correlations, while overlooking the underlying

Zhengbo Yang, Chao He, Yu Liu, Sai Wang

Pluto possesses a thin atmosphere primarily composed of N2, with minor constituents including CO and CH4. Photochemical processes generate distinct haze layers as observed by the New Horizons spacecraft. However, the mechanisms governing haze formation, as well as the composition and physical properties of the hazes, remain poorly constrained. Due to Pluto's

Zhanpeng Fu, Wei Zheng, Roderich Moessner, Marin Bukov

We present a strategy for the quantum simulation of many-body lattice models with constrained Hilbert spaces. We focus on lattice gauge theories (LGTs), which underlie a wide range of phenomena in particle physics, condensed matter, and quantum information. In present-day quantum computing platforms, perfect restrictions of the Hilbert space to the desired g

Ning Jiang, Rongli Mo

We prove the quantitative propagation of chaos for stochastic particle systems with interaction in both the drift and the diffusion coefficients, provided the drift kernel is bounded and free of Lipschitz or smoothness assumptions. Our proof is based on the relative entropy framework of Jabin and Wang \cite{JW2018}, and applies and extends their work on the

Dawei Guan, Di Yang, Chengjie Jin, Jiangtao Wang

Text-to-motion generation is driven by learning motion representations for semantic alignment with language. Existing methods rely on either continuous or discrete motion representations. However, continuous representations entangle semantics with dynamics, while discrete representations lose fine-grained motion details. In this context, we propose FlowCoMot

Yakun Yu, Ashley Wiens, Adrián Barahona-Ríos, Benedict Wilkins

Visual glitches in video games degrade player experience and perceived quality, yet manual quality assurance cannot scale to the growing test surface of modern game development. Prior automation efforts, particularly those using vision-language models (VLMs), largely operate on single frames or rely on limited video-level baselines that struggle under realis

Mingyang Li, Brian Lee, Rui Zuo, Brent Bacchus

High-definition (HD) mapping tasks, which perform lane detections and predictions, are extremely challenging due to non-ideal conditions such as view occlusions, distant lane visibility, and adverse weather conditions. Those conditions often result in compromised lane detection accuracy and reduced reliability within autonomous driving systems. To address th

Junda Zhou, Zhenyang Huang, Di-Chang Chen, Jianheng Guo

GJ~486b is a close-in planet orbiting an M dwarf and is therefore expected to have undergone strong atmospheric escape. Motivated by theoretical and observational studies on the constraints of its water and atmosphere, we investigate which combinations of an primordial hydrogen-rich atmosphere and water inventory could fit the current water content implied b

Cheng Meng, Wenxin Le, Xinyi Li, Qiuyun Wang

Existing methods for detection rule generation are tightly coupled to specific input-output combinations, requiring dedicated pipelines for each. We formalize this problem as a unified mapping f:C*L->R and characterize optimal rules through semantic distance. We propose UniRule, an agentic RAG framework built on dual semantic projection spaces: detection int

Chen Huang, Zitan Jiang, Changyi Zou, Wenqiang Lei

Customer service chatbots are increasingly expected to serve not merely as reactive support tools for users, but as strategic interfaces for harvesting high-value information and business intelligence. In response, we make three main contributions. 1) We introduce and define a novel task of Proactive Information Probing, which optimizes when to probe users f

Matthias Baur, Simon Larson

We study asymptotic shape optimization for Riesz means of Robin Laplacian eigenvalues among cuboids of fixed measure. Our focus is the regime where the Robin parameter is proportional to the square root of the spectral parameter defining the Riesz means. Here, a transition emerges based on the precise ratio between the two parameters: as the spectral paramet

Xiang Li, Yifan Jiang, Yuan Wan, Xuerong Liu

Resonant inelastic X-ray scattering measurements reveal robust magnetic excitations in the perovskite iridates SrIr$_{1-x}$Sn$_x$O$_3$. We analyzed the dispersions of the magnetic excitation with spin-dilution ratio $x$ = 0, 0.03, 0.06, 0.1, and 0.2, crossing from semi-metal to spin-diluted while antiferromagnetically ordered insulators. The extracted effect

Hari Borutta, Andreas Feuerpfeil, Yasir Iqbal

The Calogero-Sutherland model is a paradigmatic integrable system describing one-dimensional non-relativistic particles with inverse-square-type interactions. At interaction strength $λ=2$, the CSM exhibits a deep connection to anyon physics, featuring the Laughlin-Jastrow polynomial as its exact ground state. Motivated by this structure, we develop a genera

Tao Zhang, Zhiguo Hao, Songhao Yang, Hongyue Ma

Considerable efforts have been made to analyze the small-signal stability of doubly fed induction generator (DFIG) systems. However, commercial confidentiality and frequency coupling make the DFIG system a grey-box multiple-input-multiple-output (MIMO) system with highly challenging stability analysis. This paper proposes an Argument-principle based stabilit

Ziqian Zhong, Shashwat Saxena, Aditi Raghunathan

Existing approaches to monitoring AI agents rely on supervised evaluation: human-written rules or LLM-based judges that check for known failure modes. However, novel misbehaviors may fall outside predefined categories entirely and LLM-based judges can be unreliable. To address this, we formulate unsupervised monitoring, drawing an analogy to unsupervised lea

Seulki Lee

Current approaches to AI safety define red lines at the case level: specific prompts, specific outputs, specific harms. This paper argues that red lines can be set more fundamentally -- at the level of value, evidence, and source hierarchies that govern AI reasoning. Using the PRISM (Profile-based Reasoning Integrity Stack Measurement) framework, we define a

Arafat Al-Dweik, Alok Kumar Shukla, Sami Muhaidat

This work derives the exact outage probability (OP) and ergodic capacity (EC) for the near user (NU) in the widely adopted two-user downlink non-orthogonal multiple access (NOMA) over fading channels. By noting that the noise and fading become dependent after successive interference cancellation (SIC), the exact analysis is derived by considering the joint p

Waqas Ahmed, Saleh O. Allehabi, Mansoor Ur Rehman

We investigate a realistic non-supersymmetric hybrid inflation model incorporating right-handed neutrinos and assess its viability in light of recent cosmological observations. At tree level, the inflaton potential yields a blue-tilted scalar spectrum, which is disfavored by current data from Planck and ACT that instead support a red tilt. We show that inclu

Yoonsu Kim, Chanbin Park, Kihoon Son, Saelyne Yang

During complex knowledge work, people engage in iterative sensemaking: interpreting information, connecting ideas, and refining their understanding. Yet in current human-AI collaboration, these cognitive processes are difficult to share and organize for AI. They arise in situ and are rarely captured without interrupting the task, and even when expressed, rem

Haq Nawaz Malik, Nahfid Nissar

We present KS-PRET-5M, the largest publicly available pretraining dataset for the Kashmiri language, comprising 5,090,244 (5.09M) words, 27,692,959 (27.6M) characters, and a vocabulary of 295,433 (295.4K) unique word types. We assembled the dataset from two source classes: digitized archival and literary material, encompassing literature, news, biographies,

Seulki Lee

AI systems increasingly shape high-stakes decisions in healthcare, law, defense, and education, yet existing governance paradigms -- AI Ethics, AI Safety, and AI Alignment -- share a common limitation: they evaluate outcomes rather than verifying the reasoning process itself. This paper introduces AI Integrity, a concept defined as a state in which the Autho

Wei Li, Hangjie Yuan, Zixiang Zhao, Borui Kang

Continual Learning (CL) aims to train neural networks on a dynamic stream of tasks without forgetting previously learned knowledge. Among optimization-based approaches, C-Flat has emerged as a promising solution due to its plug-and-play nature and its ability to encourage uniformly low-loss regions for both new and old tasks. However, C-Flat requires three a

Ziqian Zhong, Aashiq Muhamed, Mona T. Diab, Virginia Smith

Mechanistic interpretability is often motivated for alignment auditing, where a model's verbal explanations can be absent, incomplete, or misleading. Yet many evaluations do not control whether black-box prompting alone can recover the target behavior, so apparent gains from white-box tools may reflect elicitation rather than internal signal; we call this th

Saikat Sur

We derive an upper bound on the maximum balanced bipartite entanglement entropy of ground states of many-body Hamiltonians defined on a graph, agnostic to any particular model, that possesses a nontrivial automorphism group. We show that the entropy is bounded by the logarithm of a weighted sum of multiplicities of irreducible representations of the bipartit

Xiang Ren, Jin-Yu Hu, Hao-jie Xu, Shi Pu

We investigate the nonlinear response of flow harmonics $v_2,v_4$ to initial-state eccentricities $\epsilon_2,\epsilon_4$ within the Gubser-flow framework. By extending the perturbative solutions of Gubser flow, we derive analytic nonlinear response relations connecting the eccentricities $\epsilon_2,\epsilon_4$ to the flow harmonics $v_2,v_4$. Our results r

Jaehyup Seong, Juhwan Lee, Jungwoo Lee, Sean Kwon

Dual-polarized transmission offers a promising approach to improve spectral efficiency in multiantenna networks by reusing frequency and time resources across orthogonal polarization domains. Building upon this advantage, this paper investigates interference management in mixed dual-polarized integrated satellite-terrestrial networks (MDP-ISTN), comprising a

G. E. Volovik

In the fermionic liquids, the Fermi surface is topologically stable,\cite{Volovik2003} which is at the origin of the applicability of the Landau theory of Fermi liquid (LFL). The LFL exists under special condition, when the Green's function has a pole with nonzero residue $Z$. Otherwise one has non-Landau Fermi liquid (NLFL), such as Luttinger liquid, which

Jiaxiang Wang, Zhaohui Yang, Yahao Ding, Ye Hu

Task-oriented semantic communication emerges as a crucial paradigm for next-generation wireless networks, aiming to efficiently transmit task-relevant information while reducing interference and redundancy across multiple users. Existing information bottleneck (IB)-based frameworks predominantly focus on single-user scenarios, neglecting cross-user semantic

Sergei Barakin, Kirill Gubarev, Edvard T. Musaev

We investigate uni-vector deformation in the Type IIA setup and show that the D0-brane background is mapped into itself (sedimentation), and other extremal backgrounds get bound with a dissolved D0-brane charge. Explicitly we generate F1-D0 and D2-D0 bound states background from uni-vector deformations. For the former we show that deformation of the non-extr

Jihoon Jeong

We extract 21-emotion vector sets from twelve small language models (six architectures x base/instruct, 1B-8B parameters) under a unified comprehension-mode pipeline at fp16 precision, and compare the resulting geometries via representational similarity analysis on raw cosine RDMs. The five mature architectures (Qwen 2.5 1.5B, SmolLM2 1.7B, Llama 3.2 3B, Mis

Alexander Hazeltine, Chi-Heng Lo

We give an algorithm to compute the Pyasetskii involution for $\mathrm{Sp}_{2n}$, $\mathrm{SO}_{2n+1}$ and $\mathrm{O}_{2n}$. The algorithm is a combination of Moeglin-Waldspurger's algorithm for the Pyasetskii involution for $\mathrm{GL}_n$ ([MW86]) and Lanard-M${\'i}$nguez's algorithm for the Aubert-Zelevinsky involution of bad parity representations for c

Ritu Gupta, Catherine Witteveen, Debarchan Das, Fabian O. von Rohr

We report on the microscopic superconducting properties of the Dirac semimetal PdTe2. In this study, we have focused on mosaic crystals of PdTe2, and used detailed zero field and transverse field muon spin relaxation/rotation ($\mu$SR), ac-magnetic susceptibility, and resistivity measurements to investigate their superconducting properties. The magnetic susc

Huacan Wang, Jie Zhou, Ningyan Zhu, Shuo Zhang

AI coding agents have become central to developer workflows, yet every existing solution locks its reasoning capabilities within a specific delivery form, such as a CLI, IDE plugin, or web application. This limitation creates systemic barriers when enterprises attempt to reuse these capabilities across heterogeneous engineering environments. To address this

Lik Tung Fu, Qihang Wang, Shaokai Ren, Mengli Zhang

Functional verification remains a dominant cost in modern IC development, and SystemVerilog Assertions (SVAs) are critical for simulation-based monitoring and formal property checking. However, writing SVAs by hand is time-consuming and error-prone. Directly prompting general-purpose large language models (LLMs) is also unreliable: the generated properties a

Renyu Li, Vladimir Kirilenko, Yao You, Crag Wolfe

Fine-tuning object detection (OD) models on combined datasets assumes annotation compatibility, yet datasets often encode conflicting spatial definitions for semantically equivalent categories. We propose an agentic label harmonization workflow that uses a vision-language model to reconcile both category semantics and bounding box granularity across heteroge

Jia Luo

Semiconductor supply chains face unprecedented resilience challenges amidst global geopolitical turbulence. Conventional Large Language Model (LLM) planners, when confronting such non-stationary "Policy Black Swan" events, frequently suffer from Decision Paralysis or a severe Grounding Gap due to the absence of physical environmental modeling. This paper int

Dugang Liu, Zulong Chen, Chuanfei Xu, Jiaxuan He

Office automation (OA) systems play a crucial role in enterprise operations and management, with access control flow approval (ACFA) being a key component that manages the accessibility of various resources. However, traditional ACFA requires approval from the person in charge at each step, which consumes a significant amount of manpower and time. Its intell

Qingxia Feng, Pan Fang, Meng Hua, Chunguo Li

Extremely large-scale multiple-input multiple-output (XL-MIMO) is a key enabler for sixth-generation (6G) communications. However, near-field channel estimation is particularly challenging due to spherical-wave propagation and spatial non-stationarity. To tackle this challenge, we propose a structured sparse Bayesian learning framework with adaptive dictiona

Weikun Peng, Denys Iliash, Manolis Savva

We present EgoFun3D, a coordinated task formulation, dataset, and benchmark for modeling interactive 3D objects from egocentric videos. Interactive objects are of high interest for embodied AI but scarce, making modeling from readily available real-world videos valuable. Our task focuses on obtaining simulation-ready interactive 3D objects from egocentric vi

Tao Wang, Suhang Zheng, Xiaoxiao Xu

Multi-step agentic reinforcement learning benefits from fine-grained credit assignment, yet existing approaches offer limited options: critic-free methods like GRPO assign a uniform advantage to every action in a trajectory, while learned value networks introduce notable overhead and can be fragile under sparse rewards. We observe that group rollouts targeti

Ashwin Vinod, Katrin Erk

Scientific fact-checking is vital for assessing claims in specialized domains such as biomedicine and materials science, yet existing systems often hallucinate or apply inconsistent reasoning, especially when verifying technical, compositional claims against an evidence snippet under source and cost/latency constraints. We present a pipeline centered on atom

Yifan Yu, Yuqing Jian, Junxiong Wang, Zhongzhu Zhou

Diffusion language models promise parallel generation, yet still lag behind autoregressive (AR) models in quality. We stem this gap to a failure of introspective consistency: AR models agree with their own generations, while DLMs often do not. We define the introspective acceptance rate, which measures whether a model accepts its previously generated tokens.

Rahul K. Singh, Marco E. Rosti

Turbulent polymeric flows show strong deviations from Kolomogorov-like behaviour resulting from more complex dynamics compared to Newtonian turbulence. We now study the nature of mixing in polymeric turbulence via Eulerian passive scalar fields of varying molecular diffusivities, given by the Schmidt number Sc. We show that polymeric turbulence is a less eff

Salvatore Flavio Pileggi

AI systems are consistently evolving in terms of both capability and autonomy with an holistic social impact. In this context of proliferation and fast technological evolution, the scientific community is actively engaged to assure Trustworthy AI. However, in general terms, AI safety research is significantly slower and is facing critical challenges in terms

Zhu Xu, Xun Shi

Radio pulsars, as highly coherent point sources, serve as powerful probes of the ionized interstellar medium (IISM). Pulsar scintillation observations have revealed inverted arclets on the secondary spectrum, indicating quasilinearly aligned images created by substructures on a scintillation screen. The density profiles of these substructures remain unconstr

Yassine El Gantouh, Jun Zheng, Guchuan Zhu, Dingshi Li

We study the input-to-state stability (ISS) of boundary control systems allowing for infinitely many boundary couplings. Using semigroup perturbation theory and the theory of positive linear operators on Banach lattices, we derive a spectral small-gain condition ensuring exponential ISS. We further investigate linear Boltzmann-type equations on an infinite n

Yanyan Song, Yaping Mao

For an integer $t \geq 3$, let $\mathcal{L}(t)$ denote the linear equation $x_1 + x_2 + \cdots + x_{t-1} = x_t,$ where all variables are positive integers. For integers $k \geq 1$ and $t_0,t_1,\dots,t_{k-1} \geq 3$, the generalized Schur number $S(k;t_0,t_1,\dots,t_{k-1})$ is the least positive integer $N$ such that every $k$-coloring of $[1,N]$, for some $i

Zachary Kincaid, Shaowei Zhu

Users of program analyses expect that results change predictably in response to changes in their programs, but many analyses fail to provide such robustness. This paper introduces a theoretical framework that provides a unified language to articulate robustness properties. By modeling programs and their properties as objects in a category, diverse notions of

Amir K. Goharshady, Kerim Kochekov, Tian Shu, Ahmed Khaled Zaher

Binary size reduction is an increasingly important optimization objective for compilers. One emerging technique is function merging, where multiple similar functions are merged into one, thereby eliminating redundancy. The SOTA approach to perform the merging is based on sequence alignment, where functions are viewed as linear sequences of instructions that

Jialu Pan, Yufeng Zhang, Nan Hu, Zhenbang Chen

We study the problem of characterizing the stability of Kullback-Leibler (KL) divergence under Gaussian perturbations beyond Gaussian families. Existing relaxed triangle inequalities for KL divergence critically rely on the assumption that all involved distributions are Gaussian, which limits their applicability in modern applications such as out-of-distribu

Mahdieh Zaker, Andrii Mironchenko, Amy Nejati, Abolfazl Lavaei

This paper develops a direct data-driven framework for infinite networks with unknown nonlinear polynomial subsystems, enabling the synthesis of controllers that ensure the entire network is uniformly globally asymptotically stable (UGAS). To address scalability challenges arising from high dimensionality, we develop a data-driven approach to construct an in

Nouhaila Innan, Rachmad Vidya Wicaksana Putra, Muhammad Shafique

Most quantum machine learning (QML) pipelines still rely on static encodings such as angle and amplitude maps, and this limits their ability to handle temporal information. To address this limitation, this paper uses spike-based data representation as an effective encoding mechanism that incorporates temporal structure into quantum feature preparation. Speci

Gregory Morse, Tamás Kozsik

We study when a programming language can emulate programs written in that same language without delegating the guest program back to the host evaluator or compiler. We call this property emulation-completeness. The central observation is that Turing-completeness by itself is not enough: a self-emulator must not only compute the guest program's result, but mu

Jack Kolb, Aditya Garg, Nikolai Warner, Karen M. Feigh

We investigate estimating a human's world belief state using a robot's observations in a dynamic, 3D, and partially observable environment. The methods are grounded in mental model theory, which posits that human decision making, contextual reasoning, situation awareness, and behavior planning draw from an internal simulation or world belief state. When in t

Kotaro Kikuchi, Nami Ogawa

Professional designers work from client briefs that specify goals and constraints but often lack concrete design details. Translating these abstract requirements into visual designs poses a central challenge, yet existing tools address specific aspects or induce fixation through complete outputs. Through interviews with six professional designers, we identif

Meng Yuan, Tianyou Chai

Precision contouring control is crucial in industrial machining processes, particularly for applications such as laser and water jet cutting, where contouring accuracy directly determines product quality. This paper presents a novel control strategy for biaxial machines featuring position-dependent flexibility and input delays, ensuring that the end-effector

Chamath Wanigasooriya, Indrajith Ekanayake

Cloud native architecture is about building and running scalable microservice applications to take full advantage of the cloud environments. Managed Kubernetes is the powerhouse orchestrating cloud native applications with elastic scaling. However, traditional Kubernetes autoscalers are reactive, meaning the scaling controllers adjust resources only after th

Liangqi Liu, Wenqiang Pu, Yingru Li, Zhi-Quan Luo

The dynamic competition against intelligent jammer systems presents a significant challenge to modern radar. Traditional active anti-jamming strategy learning methods often suffer from low sample efficiency and fail to fully exploit the structures of the adversary jammer. To reveal the inherent structure, this paper adopts an Online Convex Optimization (OCO)

Joel M. Sunil, J. Bharathi Kannan, Monu Bhartiya, Rayees A S

Lattice quasi-periodicity is easily realized with ultracold atoms in optical lattices and has been used to study delocalization-localization transition at low dimensions. Models with true disorder, however, remains largely unrealized in experiments. Here, using Bose-Einstein Condensate of ${^{87}{\text{Rb}}}$ atoms, we realize a Generalized Aubry-Andr\'e (GA

Chihiro Matsui

We construct a class of quantum many-body systems hosting an $\mathfrak{su}(3)$-invariant scar subspace, extending the conventional paradigm of quantum many-body scars beyond equally spaced spectra and single-directional tower structures. Our construction is based on local constraints that realize an algebraic closure within the scar subspace. As a result, t

Weiyuan He, Chen Wu, Pengwen Dai, Wei Wang

Ultra-high-definition (UHD) video denoising requires simultaneously suppressing complex spatio-temporal degradations, preserving fine textures and chromatic stability, and maintaining efficient full-resolution 4K deployment. In this paper, we propose UHD-GPGNet, a Gaussian-process-guided local spatio-temporal denoising framework that addresses these requirem

Vinooth Kulkarni, Aaron Orenstein, Xinpeng Li, Shuai Xu

The quantum computing community is increasingly positioning quantum processors as accelerators within classical HPC workflows, analogous to GPUs and TPUs. However, many real-world applications require scaling to hundreds or thousands of physical qubits to realize logical qubits via error correction. To reach these scales, hardware vendors employing diverse t

Yuanhao Ding, Meimingwei Li, Esteban Garces Arias, Matthias Aßenmacher

The quality of text generated by large language models depends critically on the decoding sampling strategy. While mainstream methods such as Top-$k$, Top-$p$, and Min-$p$ achieve a balance between diversity and accuracy through probability-space truncation, they share an inherent limitation: extreme sensitivity to the temperature parameter. Recent logit-spa

Jon-Paul Cacioli

We present this as a negative result with an explanatory mechanism, not as a formal upper bound. Predictive coding networks (PCNs) admit a K-way energy probe in which each candidate class is fixed as a target, inference is run to settling, and the per-hypothesis settled energies are compared. The probe appears to read a richer signal source than softmax, sin

Taketo Tominaga, Ryo Onishi

Multiphase turbulent flow phenomena are observed not only in industrial devices but also in environmental flows, and direct numerical simulation (DNS) plays a key role in their investigation. Many numerical models have been developed; nevertheless, few models are highly optimized for GPU platforms, which represent the current mainstream in high-performance c

Takahiko Furuya

Semantic segmentation of 3D point cloud scenes is a crucial task for various applications. In real-world scenarios, training segmentation models often faces three concurrent forms of data insufficiency: scarcity of training scenes, scarcity of point-level annotations, and absence of 2D image sequences from which point clouds were reconstructed. Existing data

Zhiyuan Zhang, Zijian Zhou, Linjun Li, Long Chen

3D texture generation is receiving increasing attention, as it enables the creation of realistic and aesthetic texture materials for untextured 3D meshes. However, existing 3D texture generation methods are limited to producing only a few types of non-emissive PBR materials (e.g., albedo, metallic maps and roughness maps), making them difficult to replicate

Haomin Zuo, Yidi Li, Luoxiao Yang, Xiaofeng Zhang

While diffusion Multimodal Large Language Models (dMLLMs) have recently achieved remarkable strides in multimodal generation, the development of interpretability mechanisms has lagged behind their architectural evolution. Unlike traditional autoregressive models that produce sequential activations, diffusion-based architectures generate tokens via parallel d

Muhammad Kamran Janjua, Abdul Wahab, Bahador Rashidi

In this work, we introduce a new perspective on comparative image assessment by representing an image pair as a structured composition of its regions. In contrast, existing methods focus on whole image analysis, while implicitly relying on region-level understanding. We extend the intra-image notion of a scene graph to inter-image, and propose a novel task o