LLM Agent Training

LLM: Large Language Model

LLM agent training studies how to make language models act reliably over multiple steps, using tools, software, websites, or simulated worlds to complete tasks. Common approaches combine supervised fine-tuning on interaction traces with reinforcement learning from execution-verified outcomes, alongside distillation, preference learning, and feedback on individual steps. Researchers build executable environments and benchmarks such as SWE-bench, WebArena, Terminal-Bench, and τ²-Bench to provide repeatable practice and test tool use, coding, search, and long-horizon workflows. A central challenge is transferring gains beyond training environments while balancing task success with efficiency, safety, and robustness to changing tools and users.

Latest papers

590 papers130 papers in the last four weeks, up 132% on the four weeks before. 0.8% of all new papers.

SWE-Fuse: Empowering Software Agents via Issue-free Trajectory Learning and Entropy-aware RLVR Training
Mar 9, 2026Xin-Cheng Wen, Binbin Chen, Haoxuan Lan +3Software Engineering Agents LLM Agent Training
Ant Group · The Chinese University of Hong Kong
SynPlanResearch-R1: Encouraging Tool Exploration for Deep Research with Synthetic Plans
Mar 9, 2026Hansi Zeng, Zoey Li, Yifan Gao +7Synthetic Data Generation for LLM Agents LLM Agent Training
University of Massachusetts Amherst · Amazon · University of Montreal +2
The Trace Is the State: Exact Credit Assignment for LLM Agent Teams
Mar 6, 2026Yanjun Chen, Yirong Sun, Hanlin Wang +5Credit Assignment Counterfactual Credit Assignment
Eastern Institute of Technology · The Hong Kong Polytechnic University · Harbin Institute of Technology
KARL: Knowledge Agents via Reinforcement Learning
Mar 5, 2026Jonathan D. Chang, Andrew Drozdov, Shubham Toshniwal +23Synthetic Data Generation for LLM Agents LLM Agent Training
Databricks AI Research
WebFactory: Automated Compression of Foundational Language Intelligence into Grounded Web Agents
Mar 5, 2026Sicheng Fan, Qingyun Shi, Shengze Xu +5Web Agents LLM Agent Training
Fudan University · IMean AI · The Chinese University of Hong Kong +1
Contextualized Privacy Defense for LLM Agents
Mar 3, 2026Yule Wen, Yanzhe Zhang, Jianxun Lian +3Privacy Protection for LLM Agents LLM Agent Security
Tsinghua University · Georgia Tech · Microsoft +1
AOI: Turning Failed Trajectories into Training Signals for Autonomous Cloud Diagnosis
Mar 3, 2026Pei Yang, Wanyi Chen, Asuka Yuxi Zheng +13LLM Agent Training AI Agent Reliability
Gradient · Soochow University · UC Santa Cruz +4
Expanding LLM Agent Boundaries with Strategy-Guided Exploration
Mar 2, 2026Andrew Szot, Michael Kirchhof, Omar Attia +1RL Exploration LLM Agents
Apple
CoVe: Training Interactive Tool-Use Agents via Constraint-Guided Verification
Mar 2, 2026Jinpeng Chen, Cheng Gong, Hanbo Li +9Tool-Using Agents LLM Agent Training
Huawei Research · Independent Researcher
TopoCurate:Modeling Interaction Topology for Tool-Use Agent Training
Mar 2, 2026Jinluan Yang, Yuxin Liu, Zhengyu Chen +7LLM Agent Training RL for Tool Use
Zhejiang University, Hangzhou, China · Meituan, Beijing, China.
FT-Dojo: Towards Autonomous LLM Fine-Tuning with Language Agents
Mar 2, 2026Qizheng Li, Yifei Zhang, Xiao Yang +4LLM Fine-Tuning AI Agent Benchmarks
Microsoft Research Asia · Peking University · Nanjing University +1
SciDER: Scientific Data-centric End-to-end Researcher
Mar 2, 2026Ke Lin, Owais Aijaz, Yilin Lu +3AI Agents for Scientific Discovery Scientific Workflow Automation
Data Science, College of William & Mary, Williamsburg, USA · Natural Language Processing, MBZUAI, Abu Dhabi, UAE · Computer Science, University of Minnesota, Minneapolis, USA +1
Evaluating Agents Across Runtime Contracts: When Mismatch Costs Efficiency or Quality
Mar 1, 2026Victor May, Van Khue Nguyen, Aaditya Salgarkar +3Tool-Using Agents LLM Agent Evaluation
Ontocord · École Polytechnique · Carnegie Mellon University +3
DeepResearch-9K: A Challenging Benchmark Dataset of Deep-Research Agent
Mar 1, 2026Tongzhou Wu, Yuhao Wang, Xinyu Ma +4Deep Research Agents AI Agent Benchmarks
Shenzhen Technology University Shenzhen, China · City University of Hong Kong Hong Kong, China · Baidu Inc. Beijing, China
Qwen3-Coder-Next Technical Report
Feb 28, 2026Ruisheng Cao, Mouxiang Chen, Jiawei Chen +17Code Language Models AI Coding Agents
InfoPO: Information-Driven Policy Optimization for User-Centric Agents
Feb 28, 2026Fanqi Kong, Jiayi Zhang, Mingyi Deng +3LLM Agent Training RL for LLM Agents
Peking University · DeepWisdom · The Hong Kong University of Science and Technology (Guangzhou) +1
LOGIGEN: Logic-Driven Generation of Verifiable Agentic Tasks
Feb 28, 2026Yucheng Zeng, Weipeng Lu, Linyun Liu +9Synthetic Data Generation for LLM Agents LLM Agent Trajectory Synthesis
Baidu Inc. · Tsinghua University
SWE-rebench V2: Language-Agnostic SWE Task Collection at Scale
Feb 27, 2026Ibragim Badertdinov, Maksim Nekrashevich, Anton Shevtsov +1Software Engineering Benchmarks Software Engineering Agents
Nebius, Amsterdam, Netherlands.
ProductResearch: Training E-Commerce Deep Research Agents via Multi-Agent Synthetic Trajectory Distillation
Feb 27, 2026Jiangyuan Wang, Kejun Xiao, Huaipeng Zhao +2LLM Agent Training LLM Agent Trajectory Synthesis
Alibaba International Digital Commercial Group
Search More, Think Less: Rethinking Long-Horizon Agentic Search for Efficiency and Generalization
Feb 26, 2026Qianben Chen, Tianrui Qin, King Zhu +21Agentic Search Long-Horizon LLM Agents
Search-P1: Path-Centric Reward Shaping for Stable and Efficient Agentic RAG Training
Feb 26, 2026Tianle Xia, Ming Xu, Lingxiang Hu +7RL for RAG Agentic RAG
Tencent
Training Agents to Self-Report Misbehavior
Feb 25, 2026Bruce W. Lee, Chen Yueh-Han, Tomek KorbakAI Agent Safety LLM Agent Training
UPenn · MATS · NYU +1
ARLArena: A Unified Framework for Stable Agentic Reinforcement Learning
Feb 25, 2026Xiaoxuan Wang, Han Zhang, Haixin Wang +11Agentic RL Reinforcement Learning
Department of Computer Science, University of California, Los Angeles, California, USA · University of Wisconsin–Madison, Madison, Wisconsin, USA
Tool-R0: Self-Evolving LLM Agents for Tool-Learning from Zero Data
Feb 24, 2026Emre Can Acikgoz, Cheng Qian, Jonas Hübotter +3LLM Agent Training Self-Evolving Agents
UIUC · ETH Zurich
On Data Engineering for Scaling LLM Terminal Capabilities
Feb 24, 2026Renjie Pi, Grace Lam, Mohammad Shoeybi +3Terminal Agents LLM Agent Training
Learning to Rewrite Tool Descriptions for Reliable LLM-Agent Tool Use
Feb 23, 2026Ruocheng Guo, Kaiwen Dong, Xiang Gao +1LLM Tool Use LLM Agent Training
Intuit AI Research, Mountain View, CA, USA
ComplLLM: Fine-tuning LLMs to Discover Complementary Signals for Decision-making
Feb 23, 2026Ziyang Guo, Yifan Wu, Jason Hartline +2LLM Decision-Making LLM Agent Training
Northwestern University · Microsoft Research · Carnegie Mellon University
Proximity-Based Multi-Turn Optimization: Practical Credit Assignment for LLM Agent Training
Feb 22, 2026Yangyi Fang, Jiaye Lin, Xiaoliang Fu +4LLM Agent Training RL for LLM Agents
Tsinghua University · Fudan University · Peking University +3
MagicAgent: Towards Generalized Agent Planning
Feb 22, 2026Xuhui Ren, Shaokang Dong, Chen Yang +21LLM Agent Planning LLM Agent Training
Honor Device Co., Ltd · Fudan University
DeepInnovator: Triggering the Innovative Capabilities of LLMs
Feb 21, 2026Tianyu Fan, Fengji Zhang, Yuxiang Zheng +5Research Idea Generation LLM Agent Training
♣The University of Hong Kong · ♠City University of Hong Kong · ♢Shanghai Jiao Tong University +1

Mar 9, 2026J/K move · Enter open · S save

SWE-Fuse: Empowering Software Agents via Issue-free Trajectory Learning and Entropy-aware RLVR Training

Xin-Cheng Wen, Binbin Chen, Haoxuan Lan, Hang Yu+2

Ant Group · The Chinese University of Hong Kong

Software Engineering Agents LLM Agent Training Automated Program Repair RL for Code Generation Reinforcement Learning with Verifiable Rewards

Open paper AskPDF arXiv

Large language models (LLMs) have transformed the software engineering landscape. Recently, numerous LLM-based agents have been developed to address real-world software issue fixing tasks. Despite their state-of-the-art performance, Despite achieving state-of-the-art performance, these agents face a significant challenge: Insufficient high-quality issue descriptions. Real-world datasets often exhibit misalignments between issue descriptions and their corresponding solutions, introducing noise and ambiguity that mislead automated agents and limit their problem-solving effectiveness. We propose SWE-Fuse, an issue-description-aware training framework that fuses issue-description-guided and issue-free samples for training SWE agents. It consists of two key modules: (1) An issue-free-driven trajectory learning module for mitigating potentially misleading issue descriptions while enabling the model to learn step-by-step debugging processes; and (2) An entropy-aware RLVR training module, which adaptively adjusts training dynamics through entropy-driven clipping. It applies relaxed clipping under high entropy to encourage exploration, and stricter clipping under low entropy to ensure training stability. We evaluate SWE-Fuse on the widely studied SWE-bench Verified benchmark shows to demonstrate its effectiveness in solving real-world software problems. Specifically, SWE-Fuse outperforms the best 8B and 32B baselines by 43.0% and 60.2% in solve rate, respectively. Furthermore, integrating SWE-Fuse with test-time scaling (TTS) enables further performance improvements, achieving solve rates of 49.8% and 65.2% under TTS@8 for the 8B and 32B models, respectively.