cs.AI · cs.CL · cs.LG · cs.CV · cs.CR · cs.SE · stat.ML · q-fin.TR · q-fin.ST · econ.GN · q-bio.QM · physics.soc-ph · math.OC · cs.RO · cs.HC
Zhanshuo Cao, Jiakai Tang, Yuqi Zhou, Sunhao Dai, Jun Xu · 2026-10-21
Existing sequential recommenders usually allocate a fixed computational budget to every item in a user's interaction history, regardless of the item's contribution to the final prediction. We challenge this de facto design with two diagnostic findings. (1) Non-uniform computation…
Yunhao Luo, Wentao He, Yinpei Dai, Hyelin Nam, Hongze Fu and others · 2026-10-21
Existing world models typically perceive a scene through current visual frame or a narrow history window. However, these observations may not reveal the physics properties that determine transition dynamics, such as an object's mass distribution, joint damping, or surface frictio…
Shibo Jin, Zhiqiu Xu, Yiling Zou, Shangyu Gong, Mayur Naik · 2026-10-21
Through a tangle of real-world interactions, from synthetic training pipelines to data shared on the open web, frontier LLMs absorb what their predecessors and competitors create. How well does each model play its role in this ecosystem, as a creator and as a solver? We introduce…
Clément Burgelin, Rachid Guerraoui, Zicong Hong, Yinan Hu, Anne-Marie Kermarrec and others · 2026-10-21
Consensus is a fundamental problem in distributed computing and at the core of state machine replication (SMR). Despite decades of active human research, discovering practical low-latency solutions to the problem is still open, as it requires combining protocol knowledge, correct…
Houde Qian, Zheng Jiang, Wenyi Li, Chenxi Xu, Xinjie Lin and others · 2026-10-21
Quantitative research is an iterative loop: a researcher proposes a market hypothesis, implements it as an executable signal, tests it on historical data, and decides what to keep. Existing benchmarks score formulas, code, or individual trading decisions, leaving open whether lar…
Yanzhe Chen, Ziyi Yang, Jifeng Zhu, Qiming Huang, Ruihe An and others · 2026-10-21
Multimodal agents (MMAs) sustain interaction through reasoning, memory, tools, and feedback, most visibly as computer-use agents, while robotic systems (RSs) couple sensing and actuation under physical dynamics. We define multimodal embodied agents (MMEAs) as goal-directed system…
Sriyash Poddar, Yanda Bao, Jacob Krantz, Matthew Chang, Xavier Puig and others · 2026-10-21
Sequential decision-making under partial observability is difficult because agents must explore to infer hidden, task-relevant information before acting effectively. Learning such information-gathering behaviors tabula rasa via online reinforcement learning is sample-inefficient,…
Sonja Topf, Ana Sanchez-Fernandez, Günter Klambauer · 2026-10-21
Despite recent advances, current AI systems remain highly susceptible to domain shifts, particularly in biomedical imaging, where images are usually acquired under substantially different experimental conditions. Although recent domain adaptation methods have made substantial pro…
Meng Ding, Mingxi Lei, Jie Zhang, Jinyan Liu, Di Wang · 2026-10-21
Reinforcement Learning with Human Feedback (RLHF) has become a central paradigm for aligning Large Language Models with human values. However, the reliance on sensitive preference data necessitates rigorous differential privacy (DP) guarantees to prevent user information leakage.…
Likui Zhang, Zhe Zhu, Zhihao Zhan, Yibo Bai, Hanlin Wang and others · 2026-10-21
Vision-language-action (VLA) models have shown promise in robotic manipulation, but imitation learning (IL) alone often struggles to master complex long-horizon skills. Reinforcement learning (RL) offers an effective avenue for further improving VLA policies. However, existing VL…
Hanmin Li, Yury Demidovich, Peter Richtárik · 2026-10-21
Parallel AI agents can produce useful local edits yet degrade a shared solution when their edits are combined. We model this coordination problem as discrete block optimization with an exactly evaluable objective and a known dependency graph. Independent sets certify compatible e…
Xinzhe Wang, Xudong Weng, Zang Li, Shaolei Zhang · 2026-10-21
Data agents built on ReAct-style loops now execute complex analytical workflows over structured data, yet their intermediate reasoning still runs almost entirely through text. Human analysts instead move between numbers, tables, and charts, reading visual evidence to expose trend…
May Espinola, Qifan Zhang, Yu Xiang · 2026-10-21
Recent advances in frontier AI models have demonstrated remarkable capabilities in visual understanding, reasoning, planning, and tool use, raising the possibility of using general-purpose AI models for robot manipulation without task-specific training. However, their manipulatio…
Lorenz Brehme, Benjamin Steenhoek, Ruth Breu · 2026-10-21
Coding agents are typically evaluated using one-shot, well-formulated prompts. Benchmarks such as SWE-bench use existing GitHub issues as input for a coding agent and evaluate whether the agent successfully solves the task. However, in real-world use cases, prompts are often not …
Tianyou Lai, Wentao Yue, Zhiwei Zuo, Qilei Li · 2026-10-21
Decision models map stated rules and observed states to discrete answers. Aggregate accuracy leaves unresolved which rule and information changes guide those answers. We connect controlled input changes to transfer, joint reliability, and directed information use through executab…
Zhaohui Yang, Quanbao Lin, Chengqi Shi, Zhenglang Weng, Jicheng Lin and others · 2026-10-21
Humanoid loco-manipulation requires an effective interface between high-level task policies and whole-body control. Existing interfaces typically rely on a single action representation for both task intent and whole-body coordination. This may burden the high-level policy or leav…
Sudipta Paul, Pengrui Quan, Aounon Kumar, Nicholas Tzou, Vivek Kulkarni and others · 2026-10-21
Harness optimization, which improves an agent's prompts, tools, and workflows, is an increasingly effective way to enhance agents built on frozen or closed-source LLMs. Leading approaches such as the Darwin--Godel Machine and the Huxley--Godel Machine (HGM) rely on recursive self…
Zijian Jiang, Xilu Wang · 2026-10-21
Text skills can extend the capabilities of tool-using agents, but long skill descriptions increase computational costs. Soft prompts provide a compact alternative, yet learning from entire successful trajectories may also introduce unnecessary behavioral changes. We find that the…
Yikai Zhao, Saurabh Pandey, Pradeep Kumar Misra · 2026-10-21
Enterprise agents act on documents. Before an agent can follow a policy, answer a compliance question or file a ticket, some ingestion layer must turn a Word file or PDF into tokens - and that layer is where an agent deployed in the wild silently loses information. We show that o…
Yuhao Chen, Donald G. Truhlar, Xiao He · 2026-10-21
Empirical exchange-correlation functionals are fitted on fixed orbitals but used self-consistently, so better energies often come with worse densities. We show that the Brillouin residual of a trained functional on the Kohn-Sham orbitals of an accurate density controls, to first …
Meihong Jia, Yasemin Timar, Jonathan Taws · 2026-10-21
This study introduces GuideEvolve, a self-evolving agent framework incorporating expert guidance and three autonomous regulatory mechanisms, collectively termed the 3Ts: Think Fast and Slow, Throwback, and Termination by choice. The framework addresses critical limitations in ext…
Sujoy Roychowdhury, Joseph H. R. Isaac, Jiangning Gao, Martin Klitte, Diarmuid Corcoran · 2026-10-21
Understanding large, complex, and rapidly evolving code bases remains a core challenge for software engineering at scale, hindering onboarding, maintenance, refactoring, and impact analysis. Providing this information to a Large-Language Model (LLM) is another hurdle, as the LLM …
Shota Fujikawa, Takuya Tamura, Kosuke Akimoto, Daichi Haraguchi, Masafumi Oyamada · 2026-10-21
LLM agents with similar success rates can follow different action sequences. An Agent Behavior Map represents LLM-harness combinations using centered action log-likelihoods on shared reference trajectories, without requiring new environment rollouts from the target agent. We deri…
Liam Dugan, Callum McDougall, Matthew Denton, Philip Pham, Christine Kaeser-Chen and others · 2026-10-21
AI-generated text detectors are increasingly deployed in consequential settings---peer review, education, admissions, journalism, and content moderation---where their outputs directly affect individuals. Yet the strongest detectors are fine-tuned neural classifiers whose decision…
Madhulatha Mandarapu, Sandeep Kunkunuru · 2026-10-21
Alarm root-cause analysis is studied on data that cannot be shared. Operator logs are confidential, so what the field publishes is a causal graph between alarm types rather than per-incident causes, a fixed scenario set, gated, or never released. Two recent benchmarks carry per-i…
Tanu Singh, Aashna Suman, Vinod Patidar · 2026-10-21
Natural-language software requirements often omit assumptions about performance, operating conditions, user behaviour, and domain constraints, affecting architecture, testing, and deployment. If undiscovered until implementation, resolving them is far more costly than addressing …
Jiayi Chen, Hui Li, Yujiong Shen, Qiyuan Peng, Yanxin Li and others · 2026-10-21
Benchmark performance is the dominant standard for comparing large language models (LLMs), typically summarized by a single aggregate score, such as accuracy, pass rate, or rubric score. This aggregate score provides limited information about a model's underlying capabilities, ye…
Chaoran Yuan, Fadi Kurdahi · 2026-10-21
Reliable validation of autonomous driving cyber-physical systems requires testing not only nominal driving behavior but also failures arising from faults in sensing and embedded computation. Physical road testing provides realistic system behavior but is costly and difficult to r…
My Chiffon Nguyen, Tianyi Hu, Vinod Anbalagan, Henry Gagnier, Pouya Sadeghi and others · 2026-10-21
As LLMs expand their language coverage, evaluating how well they understand and represent diverse cultures becomes increasingly important. However, many benchmarks remain English-centric, with limited cultural coverage and a focus on factual knowledge over its application in real…
M. Mikail Demir · 2026-10-21
Given the enormous number of claims to be checked, verifiable legal AI requires scalable, cost-effective models that answer the practical needs of law. System One decision models are built on large language models but, instead of generating text, return a calibrated probability f…
Nithin Sivakumaran, Ali Payani, Hyunji Lee, Mohit Bansal, Elias Stengel-Eskin · 2026-10-21
Forming an accurate theory-of-mind (ToM) is key to successful multi-agent collaboration and requires reasoning under uncertainty to capture multiple possible partner states. This is hard to do in the text representation space native to large language model (LLM) agents, which mor…
Yufeng Li, Zaijing Li, Fuchen Zhang, Zongzong Wu, Ming Zhao and others · 2026-10-21
Retrieval-augmented generation (RAG) frequently integrates knowledge graphs (KGs) for complex multi-hop question answering. However, their performance is increasingly bottlenecked by the knowledge structuring and reasoning processes. (i) Severe semantic loss: Distilling continuou…
Yifan Sun, Naicheng Yu, Jiawen Gong, Jingyan Shen, Huan Zhang · 2026-10-21
A benchmark is not merely a collection of test questions; it naturally induces an empirical measure over difficulty. In this work, we develop this measure-based view of benchmark evaluation. This perspective allows us to identify and formalize two central failure modes: measure m…
Prakhar Ganesh, Yash More, Marco Romanelli, Ferdinando Fioretto, Golnoosh Farnadi · 2026-10-21
As large language models (LLMs) continue to grow in scope and capability, privacy concerns are on the rise. Since users increasingly interact with a single general-purpose system across many distinct contexts, there is a natural tendency to pursue a one-size-fits-all solution to …
Siddhesh Pawar, Chen Cecilia Liu, Anna Korhonen, Isabelle Augenstein · 2026-10-21
As LLMs are increasingly deployed in value-sensitive contexts, aligning them with human values has emerged as a post-training objective in NLP, aimed at instilling additional values and guiding model behaviour. However, current evaluations remain narrow. Because values are interc…
Marcos Junca, Mauro Junca · 2026-10-21
Thermal cameras can watch over older people living alone without capturing faces, and multimodal large language models (LLMs) can judge a short video without task-specific training. We evaluate a production multimodal LLM, Gemini 3.8 Flash at temperature 0, as the decision stage …
Aravindan Kamatchi Sundaram, B. Pabitramohan Prusty, Rohit Batra · 2026-10-21
Over the past decade, substantial efforts have been devoted to developing machine learning (ML) methods for extracting materials data and, where possible, materials knowledge directly from the scientific literature. This review traces the evolution of these methods from heuristic…
Viral Patel, Boyin Tan, Jason Xue, Youcheng Sun · 2026-10-21
Coding agents can invest substantial effort in repository exploration and testing yet still produce unsuccessful patches. Existing online steering methods review trajectories periodically or invoke an advisor when a monitor detects behavioral problems; frequent reviews are costly…
Laya Iyer, Chen Shani · 2026-10-21
Large language models are typically trained with next-token prediction (NTP), which rewards predicting specific tokens rather than explicitly accounting for the semantic similarities between different words or expressions. Prior work has addressed this limitation by augmenting tr…
Rose Johnson, Debabrata Mandal, Zihao Zou, Jessie Wei, Anna Wirth-Singh and others · 2026-10-21
The mid-wave infrared (MWIR, 3-5 microns) is indispensable for thermal sensing, gas detection, long-range imaging, and free-space optical communications. Meta-optics offer a compact, lightweight, and potentially low-cost alternative to conventional refractive optics, making them …
GET https://store.agentexchange.work/gw/papers/latest?cat=cs.AI&limit=25 (0.003 USD per call). Catalogue