English
Related papers

Related papers: Ratchet: A Minimal Hygiene Recipe for Self-Evolvin…

200 papers

Reinforcement learning (RL) trained language model agents with tool access are increasingly deployed in coding assistants, research tools, and autonomous systems. We introduce the Reward Hacking Benchmark (RHB), a suite of multi-step tasks…

Machine Learning · Computer Science 2026-05-06 Kunvar Thaman

Large language models (LLMs) are becoming the foundation for autonomous agents that can use tools to solve complex tasks. Reinforcement learning (RL) has emerged as a common approach for injecting such agentic capabilities, but typically…

Machine Learning · Computer Science 2026-02-26 Emre Can Acikgoz , Cheng Qian , Jonas Hübotter , Heng Ji , Dilek Hakkani-Tür , Gokhan Tur

We present ReflexGrad, a dual-process architecture for within-episode failure recovery in LLM agents without demonstrations. When agents commit to a wrong approach early and exhaust the step budget, the post-failure trajectory contains the…

Machine Learning · Computer Science 2026-05-29 Ankush Kadu , Aswanth Krishnan

Improving performance on complex tasks and enabling interpretable decision making in large language models (LLMs), especially for clinical applications, requires effective reasoning. Yet this remains challenging without supervised…

Computation and Language · Computer Science 2025-05-26 Che Liu , Haozhe Wang , Jiazhen Pan , Zhongwei Wan , Yong Dai , Fangzhen Lin , Wenjia Bai , Daniel Rueckert , Rossella Arcucci

Large language models (LLMs) and high-capacity encoders have advanced zero and few-shot classification, but their inference cost and latency limit practical deployment. We propose training lightweight text classifiers using dynamically…

Computation and Language · Computer Science 2026-01-26 Gaurav Maheshwari , Kevin El Haddad

Agent skills today are hand-crafted, generated one-shot, or evolved through loosely controlled self-revision, none of which behaves like a deep-learning optimizer for the skill, and none of which reliably improves over its starting point…

Artificial Intelligence · Computer Science 2026-05-26 Yifan Yang , Ziyang Gong , Weiquan Huang , Qihao Yang , Ziwei Zhou , Zisu Huang , Yan Li , Xuemei Gao , Qi Dai , Bei Liu , Kai Qiu , Yuqing Yang , Dongdong Chen , Xue Yang , Chong Luo

Retrieval-augmented generation (RAG) systems empower large language models (LLMs) to access external knowledge during inference. Recent advances have enabled LLMs to act as search agents via reinforcement learning (RL), improving…

Artificial Intelligence · Computer Science 2025-11-06 Pengcheng Jiang , Xueqiang Xu , Jiacheng Lin , Jinfeng Xiao , Zifeng Wang , Jimeng Sun , Jiawei Han

We study whether self-learning can scale LLM-based agents without relying on human-curated datasets or predefined rule-based rewards. Through controlled experiments in a search-agent setting, we identify two key determinants of scalable…

Artificial Intelligence · Computer Science 2025-10-22 Wangtao Sun , Xiang Cheng , Jialin Fan , Yao Xu , Xing Yu , Shizhu He , Jun Zhao , Kang Liu

Large language models (LLMs) are increasingly used for tasks that require complex reasoning. Most benchmarks focus on final outcomes but overlook the intermediate reasoning steps - such as planning, revision, and decision making under…

Artificial Intelligence · Computer Science 2025-06-16 Xiaopeng Yuan , Xingjian Zhang , Ke Xu , Yifan Xu , Lijun Yu , Jindong Wang , Yushun Dong , Haohan Wang

As LLM agents scale to long-horizon, multi-session deployments, efficiently managing accumulated experience becomes a critical bottleneck. Agent memory systems and agent skill discovery both address this challenge -- extracting reusable…

Artificial Intelligence · Computer Science 2026-04-20 Xing Zhang , Guanghui Wang , Yanwei Cui , Wei Qiu , Ziyuan Li , Bing Zhu , Peiyang He

Existing LLM-based EDA agents are often isolated task-specific systems. This leads to repeated engineering effort and limited reuse of successful design and debugging strategies. We present LEGO, a unified skill-based platform for front-end…

Artificial Intelligence · Computer Science 2026-05-19 Jincheng Lou , Ruohan Xu , Jiecheng Ma , Runzhe Tao , Xinyu Qu , Yibo Lin

Large Language Models have evolved from single-round generators into long-horizon agents, capable of complex text synthesis scenarios. However, current evaluation frameworks lack the ability to assess the actual synthesis operations, such…

Computation and Language · Computer Science 2026-03-03 Andrew Zhuoer Feng , Cunxiang Wang , Yu Luo , Bosi Wen , Yidong Wang , Lin Fan , Yilin Zhou , Zikang Wang , Wenbo Yu , Lindong Wu , Hongning Wang , Minlie Huang

How much autonomy can multi-agent LLM systems sustain -- and what enables it? We present a 25,000-task computational experiment spanning 8 models, 4--256 agents, and 8 coordination protocols ranging from externally imposed hierarchy to…

Artificial Intelligence · Computer Science 2026-04-01 Victoria Dochkina

While current software agents powered by large language models (LLMs) and agentic reinforcement learning (RL) can boost programmer productivity, their training data (e.g., GitHub issues and pull requests) and environments (e.g.,…

Software Engineering · Computer Science 2026-05-20 Yuxiang Wei , Zhiqing Sun , Emily McMilin , Jonas Gehring , David Zhang , Gabriel Synnaeve , Daniel Fried , Lingming Zhang , Sida Wang

LLM-based agent systems increasingly rely on agent skills sourced from open registries to extend their capabilities, yet the openness of such ecosystems makes skills difficult to thoroughly vet. Existing attacks rely on injecting malicious…

Cryptography and Security · Computer Science 2026-04-08 Zenghao Duan , Yuxin Tian , Zhiyi Yin , Liang Pang , Jingcheng Deng , Zihao Wei , Shicheng Xu , Yuyao Ge , Xueqi Cheng

LLMs are powerful generators of synthetic data, which are used for training smaller, specific models. This is especially valuable for low-resource languages, where human-labelled data is scarce but LLMs can still produce high-quality text.…

Computation and Language · Computer Science 2025-10-08 Jan Cegin , Branislav Pecher , Ivan Srba , Jakub Simko

Reusable skills let LLM agents package task-specific procedures, tool affordances, and execution guidance into modular building blocks. As skill ecosystems grow to tens of thousands of entries, exposing every skill at inference time becomes…

Machine Learning · Computer Science 2026-04-02 YanZhao Zheng , ZhenTao Zhang , Chao Ma , YuanQiang Yu , JiHuai Zhu , Yong Wu , Tianze Xu , Baohua Dong , Hangcheng Zhu , Ruohui Huang , Gang Yu

Agent skills today are static artifact: authored once -- by human curation or one-shot generation from parametric knowledge -- and then consumed unchanged, with no mechanism to improve from real use. We propose \textbf{SkillEvolver}, a…

Artificial Intelligence · Computer Science 2026-05-12 Genrui Zhang , Erle Zhu , Jinfeng Zhou , Caiyan Jia , Hongning Wang

Automatic code optimization remains a difficult challenge, particularly for complex loop nests on modern hardware. This paper investigates a novel approach to code optimization where Large Language Models (LLMs) guide the process through a…

Programming Languages · Computer Science 2025-12-30 Massinissa Merouani , Islem Kara Bernou , Riyadh Baghdadi

Large language model (LLM) agents rely on reusable skills to solve complex tasks. However, existing skill creation approaches treat skills as isolated and static artifacts, limiting their reusability, reliability, and long-term improvement.…

Artificial Intelligence · Computer Science 2026-05-27 Huawei Lin , Peng Li , Jie Song , Fuxin Jiang , Tieying Zhang