OmegaUse‑SOP is a human‑in‑the‑loop system that transforms expert demonstrations of professional computer use into reusable skills for graphical user interface (GUI) agents. It refines demonstrations through four modules—Observe, Reason, Configure, and Execute—to record multimodal traces, abstract them into semantic steps, incorporate domain rules, and execute the resulting skills with verification. In a collaboration with a power‑sector client, OmegaUse‑SOP was tested on photovoltaic simulation workflows in PVsyst 7.2, showing improved reliability for GUI agents on professional SOP tasks.
By Yixiong Xiao, Lang An, Hucheng Yang, Pinxue Ma, Yongquan Chen, Jingjia Cao, Yusai Zhao, Ting Wang, Ting Liu, Siqi Bao, Jingbo Zhou, Hua Wu
The study investigates how large language models (LLMs) perform in a double auction market, a common economic mechanism. By replacing human participants with LLM agents, the authors find that markets with LLMs converge more slowly or not at all, leading to less efficient resource allocations. Analysis of trading decisions reveals significant variation across model families and roles, and a lexical study of Chain-of-Thought traces links trade execution to a shift from strategic thinking to urgency.
By Pawel Struski, Jakub Swistak, Inez Okulska, Przemyslaw Biecek
The paper presents a decision‑support system that enhances retrieval‑augmented generation (RAG) for customer contact centers by first identifying customer questions in real time. If a query matches a frequently asked question (FAQ), the system retrieves the answer directly from the FAQ database; otherwise it generates an answer via RAG, delivering responses to agents within two seconds. The approach reduces manual query formulation, lowers average handling times, and cuts operational costs, and it includes an automated workflow that uses LLMs to extract FAQs from historical transcripts when none are predefined.
By Garima Agrawal, Sashank Gummuluri, Cosimo Spera
The paper introduces a new training paradigm for text-based world models that prioritizes behavior consistency over traditional state consistency metrics. It proposes the Behavior Consistency Reward (BehR), a step-level metric that evaluates how the likelihood of a logged next action changes between real and predicted states using a frozen Reference Agent. Experiments on WebShop and TextWorld demonstrate that BehR-based training improves long-term alignment, reduces false positives in offline evaluation, and yields modest gains in lookahead planning while maintaining or enhancing single-step prediction quality.
By Youling Huang, Guanqiao Chen, Junchi Yao, Lu Wang, Fangkai Yang, Chao Du, ChenZhuo Zhao, Pu Zhao, Qingwei Lin, Saravan Rajmohan, Dongmei Zhang
The paper introduces EqGrid, a closed‑loop simulation that uses a low‑frequency, open‑weight LLM policy agent to set price, carbon limits, and subsidies for a community of empirically‑grounded household personas, while high‑frequency multi‑agent RL traders clear a continuous double auction on a physically constrained IEEE‑33‑bus grid. It demonstrates that the LLM can reduce energy‑poverty inequality—lowering the Gini of energy burden from 0.351 to 0.305 and mean burden by 28%—without increasing net grid cost, and that a compressed sub‑1B model retains 92–95% of this benefit at dramatically lower inference energy. The study also establishes a compute‑efficiency frontier and a decoupled‑safety design that eliminates grid violations.
whyItMatters":"By showing that a lightweight LLM can effectively manage energy markets to reduce poverty and inequality while staying energy‑efficient, the work offers a practical, low‑carbon AI solution for humanitarian energy‑poverty interventions."
By Kunal Jadhav, Siddhesh More
SocialBuddy is an agentic search framework designed specifically for social media scenarios, addressing challenges posed by heterogeneous user queries and multi-dimensional social feeds. It introduces SocialEnv, a large-scale simulated environment with 200K user profiles, 10 million posts, and 50K reasoning trajectories, and SocialPO, a hybrid‑granularity optimization framework that combines macro‑level reinforcement with micro‑level fine‑grained corrections to handle sparse rewards. The authors also present the SocialSearch Benchmark for evaluating social search capabilities, and report that SocialBuddy‑35B outperforms larger frontier LLMs in extensive experiments.
By Mingxuan Li, Yirong Mao, FaZhan Zhang, Haibiao Yao, Runze Hu, Wenhui Que
CivBench is an open‑source benchmark that evaluates language‑model agents in the long‑horizon, tool‑mediated game Civilization VI using the Model Context Protocol (MCP). Each episode lasts over 300 turns, generating thousands of tool calls across a 76‑tool action space, and includes a narration layer that translates visual game state into structured text. The study characterises agent behaviour across four model families, introducing Proactive Monitoring Rate (PMR) and RAG@10 as interface‑level metrics, and finds that agents often under‑monitor strategic state and fail to execute near‑term commitments despite tool access and explicit guidance.
By Austin Tudor David Andrews, Liam Wilkinson, Jamie Heagerty, Harry Coppock, Jakob Nicolaus Foerster, Rui Ponte Costa
The paper introduces a method for creating query‑specific rubrics for DeepResearch‑style long‑form report generation by training rubric generators with reinforcement learning. It builds a dataset of queries annotated with human preferences, then uses a hybrid reward that includes preference consistency, format validity, and LLM‑based rubric evaluation. The learned rubrics outperform generic or manually constructed alternatives in distinguishing preferred reports and, when used as rewards, improve performance of both single‑agent and multi‑agent DeepResearch systems.
By Changze Lv, Jie Zhou, Wentao Zhao, Jingwen Xu, Shihan Dou, Zisu Huang, Muzhao Tian, Xiaohua Wang, Zhengkang Guo, Yang Liu, Pluto Zhou, Tao Gui, Le Tian, Xiao Zhou, Xiaoqing Zheng, Xuanjing Huang, Jie Zhou
VTOS (Vision Tools Orchestration Search) is a framework that adaptively orchestrates vision foundation tools—such as open‑vocabulary detectors, segmentation models, and post‑processing operators—by jointly searching for executable solution programs and observer programs that diagnose failures and provide feedback. The observer programs feed observations into a shared VisionThoughts knowledge base, guiding subsequent searches. In two case studies—dense object counting on LVIS‑Count and zero‑shot plant‑disease segmentation on PlantSeg‑OOD—VTOS outperforms static tool pipelines and agentic visual‑programming baselines, especially in complex scenarios like dense, occluded scenes and out‑of‑distribution segmentation.
By Jinchao Ge, Lingqiao Liu, Shuwen Zhao, Lei Wang
The paper introduces CAPTURE, a system designed to help personalized language agents distinguish genuine preference changes from temporary context shifts or malicious memory poisoning. CAPTURE employs a neural differential-equation belief tracker, a multi-timescale memory ledger, uncertainty-triggered clarification, and counterfactual auditing to resolve ambiguity. Experiments on 480 episodes from 96 users show CAPTURE outperforms baseline methods, limiting poisoning success while accepting most real preference updates.
By S M Asif Hossain, Ruksat Khan Shayoni, Md Kishor Morol
The paper surveys efficient GUI agents, emphasizing that practical deployment requires more than task success—it must also minimize context, computation, action budget, and runtime overhead. It reviews key efficiency dimensions—observation, memory, action, and planner/system—and identifies recurring strategies such as selective reading, global-to-local visual allocation, recoverable memory, verification-aware control, and hybrid runtimes. The authors highlight open challenges, including accurate verifier cost accounting, benchmark comparability, and co-design of observation, memory, and execution layers under real latency and privacy constraints.
By Bizhe Bai, Jiakang Yuan, Hongming Wu, Xinyue Wang, Jie Ren, Siyao Chen, Yuchen Ya, Fan Bai, Pai Peng, Huafeng Qin, Tao Chen
The paper presents a hardware‑accelerated instance segmentation framework tailored for resource‑constrained lunar robotics, addressing low‑light perception, limited compute, and radiation‑induced hardware faults. It introduces Activation Variance Informative Sampling (AVIS), a label‑free calibration method that selects samples based on activation variance, and deploys a YOLO‑based model on a Deep Learning Processor Unit with architectural tweaks to reduce CPU fallback and ensure bounded latency. A software‑level criticality analysis estimates fault exposure, guiding mitigation that reduces global criticality by 31.7%, while AVIS with bias correction recovers 69.8% of quantization‑induced accuracy loss at 309 ms latency and 5.7 W power consumption.
By Siddhant Shete, Hilmi Dogu K\"uc\"uker, Udo Frese, Frank Kirchner
The study investigates how large language model (LLM) agents influence consensus formation in mixed human‑AI groups during a collaborative description game. Three regimes emerge: low agent proportions lead to human‑led consensus, intermediate proportions disrupt convergence, and high proportions produce strong, agent‑led consensus. The resulting consensus differs in semantic grounding and communicative form, with human‑led consensus being concrete and holistic, and agent‑led consensus being abstract and geometrically segmented.
By Lin Chen, Ziyi Liu, Xia Hu, Yong Li
CORAL is an LLM‑native harness that automates continual optimization of production recommender systems. It operates in a closed loop: an agent observes system signals, reasons over past decisions, and uses tools—including a numerical optimizer—to reconfigure the recommender while staying within a fixed operating budget. In A/B experiments on two large social platforms, CORAL improved engagement without extra serving cost on one platform and reduced serving cost without harming engagement on the other, demonstrating that a single agentic loop can replace manual engineering for ongoing system tuning.
By Muhammad Rafay Azhar, Yuhang Zhou, Gilbert Jiang, Yuchen Wang, Rahul Sharma, Matthew DeSousa, Jiayi Liu, Xin Guo, Lizhu Zhang, Xiangjun Fan
EarlyEval introduces a lightweight framework that predicts an LLM agent’s final outcome early in its execution, allowing the run to halt when a LightGBM classifier reaches a calibrated confidence threshold. By training success and failure classifiers on behavioral, textual, and reference-solution features, EarlyEval can cut 13%-26% of agent steps and up to 44.1% of input tokens while maintaining 89%-97% prediction accuracy. Across three benchmarks—SWE-bench Verified, TerminalBench, and Toolathlon—this approach reduces evaluation costs with minimal impact on per-agent resolve rates.
By Yuling Shi, Zhensu Sun, Junsen Dong, Chengcheng Wan, David Lo, Xiaodong Gu
The paper surveys the transition from textual chain-of-thought reasoning to action-grounded reasoning in autonomous driving, highlighting that driving decisions require continuous actions that mirror the spatiotemporal structure of the physical world. It reviews 171 papers, categorizing 130 methods into four main types—language-based, visual-spatial, latent-dynamic, and externalized reasoning—along with 13 subtypes linked to specific regions of interest. The authors argue that the future of driving agent reasoning lies in intermediate representations that are grounded in reality, linked to real-time actions, and verifiable within safety-critical systems.
By Zhengxu Tang, Xiaozhou Zhang, Guofeng Cui, Ziyu Gong, Zi Wang, Yunfei Shi, Ruifeng Deng, Chengzhi Qi, Ke Chen, Sachin Patil, Tianjun Xiao, Langechuan Liu, Pichao Wang
KSG‑Net introduces a Key‑Sparse and Global‑Context learning framework for maritime 3D ship detection, addressing weak feature representation of small, sparse vessels and limited global modeling of large vessels. The network employs a Key Sparse Multi‑scale Aggregation module to select informative voxels and aggregate cross‑scale features, and a Global Context Aggregation module to capture long‑range geometric dependencies via scene‑level context modeling. Experiments on the Thames River vessel dataset and simulated data show that KSG‑Net outperforms existing methods in multi‑scale vessel detection and remains robust in complex maritime environments.
By Zhouyuan Huai, Meiqi Wan, Yan Yang, Minshi Chen, Xin Yuan, Wei Wang, Xiao Wang
The paper argues that using a large language model (LLM) as the sole judge in self‑improving agent pipelines is problematic, as the judge can be biased or manipulated, leading to false confidence in system performance. The authors propose a new framework, PROCTOR, which replaces the oracle judge with a deterministic, teacher‑student loop that enforces guardrails such as sandboxing, role separation, and acceptance checks to prevent cheating and ensure reliable evaluation. Experiments across contract analysis, compliance review, and code quality demonstrate that PROCTOR mitigates eleven identified failure modes that previously allowed agents to achieve perfect scores while hiding significant capability gaps.
By Vansh Wahi
The paper introduces Repo-To-Skill, a method for converting GitHub repositories into reusable AI skills. By distilling operational knowledge from over 1,000 machine‑learning repositories, the authors build the AREX‑Skill Library with more than 5,000 verified skills across 20 areas. Integrating these skills into a research agent—DisCo—yields significant performance boosts on multiple benchmarks, demonstrating the value of reusable, task‑agnostic knowledge.
By Jianlyu Chen, Yuyang Hu, Hongjin Qian, Jiawei Liu, Wenqing Wei, Xiaolong Chen, Defu Lian, Zhicheng Dou, Chaozhuo Li, Qiwei Ye, Zheng Liu
WMLLM is a self‑evolving optimization‑agent framework that uses a predict‑then‑act world modeling approach. The agent first predicts promising optimization directions and then generates candidates, refining both its world model and strategy through multi‑turn refinement, population‑based search, and reinforcement learning. Experiments on black‑box tasks, particularly multi‑objective molecular optimization, demonstrate that WMLLM improves sample efficiency and achieves state‑of‑the‑art results within a limited evaluation budget.
By Zhongzheng Li, Qingsong Ran, Shikun Feng, Nian Ran, Wenhao Li, Xiaoyuan Zhang, Yue Wang, Xiaoguang Zhao