RL for LLM Agents

RL: Reinforcement Learning · LLM: Large Language Model

Follow this topic Ask about this topic

Momentum

70 papers in the last four weeks, up 141% on the four weeks before. 0.7% of all new papers.

Jul 13Week of Sep 28

Latest papers 402

All topics

HarnessSQL: Harness-Native Training for SQL Agents in Realistic Database Environments
Oct 8, 2026Haolin Yang, Jipeng Zhang, Jian Xie +3Text-to-SQL LLM Agent Training
Hong Kong University of Science and Technology · Microsoft Research · Tsinghua University +1
Do LLMs Learn from Rewards in Context? : Rethinking the role of reward in In-Context Reinforcement Learning
Oct 8, 2026Minchan Kwon, Seunghee Koh, Sunghyun Baek +2In-Context RL RL for LLM Agents
Korea Advanced Institute of Science and Technology (KAIST)
StoreBench: A Live-Commerce Environment for Evaluating and Training Autonomous Operator Agents
Oct 7, 2026Daksh Raghuvanshi, Ved Vedere, Yifan WangAI Agent Benchmarks Long-Horizon LLM Agents
AfterQuery
Stochastic Teacher Intervention for Agentic On-Policy Distillation
Oct 7, 2026Junnan Liu, Linhao Luo, Zhijun Chen +3On-Policy Distillation Teacher-Student Learning
Department of Data Science and AI, Faculty of Information Technology, Monash University, Australia · The Hong Kong Polytechnic University, Hong Kong SAR, China · Zhongguancun Laboratory, Beijing, P.R. China
On the Clock: Towards Punctual and Productive Time-Budgeted AI Agents
Oct 7, 2026Aaron Wang, Neelabh Madan, Vlad Sobal +5Runtime Enforcement for AI Agents LLM Agent Harnesses
University of Illinois at Chicago · AWS Agentic AI · New York University +1
UniSkill: Learning Actor-Aligned Skill Proposals for an Evolving Policy
Oct 7, 2026Yifei Lu, Cheng Liu, Dianzhi Yu +4LLM Agent Skill Learning Agent Skill Learning
Qiantang Credit, Hangzhou, China · Ant Group, Hangzhou, China · The Chinese University of Hong Kong
RewardWeaver: Long-Horizon Interactive Learning for Language Agents via Self-Evolving Reward Adaptation
Oct 7, 2026Hengbo Xiao, Boyao Zhang, Purui Liu +4RL for LLM Agents Rubric-Based RL
Learning to Accumulate Knowledge with Mutual Information
Oct 7, 2026Yuyang Zhao, Lizi Liao, Leyang Shen +4Memory-Augmented Language Model Agents RL for LLM Agents
University of Science and Technology of China · Singapore Management University · National University of Singapore
Training Advisors for LLM Agents from Task Outcomes
Oct 7, 2026Sergei Polezhaev, Barys Liskavets, Ori Press +1LLM Agent Training RL for LLM Agents
Nebius AI
SkillForge: Co-Evolving Skills and Agents via Dynamic Skill Lifecycles
Oct 7, 2026Yuyao Ge, Yiwei Wang, Yuchen He +5LLM Agent Skill Learning LLM Agents
Institute of Computing Technology, Chinese Academy of Sciences · University of California, Merced · Tsinghua University
World Potential Model: Pretrained World Knowledge as Progress Potentials
Oct 7, 2026Jun Zhao, Jixin Tang, Yang Shu +6Long-Horizon LLM Agents RL for LLM Agents
National University of Singapore · Fudan University · Tsinghua University +1
MIMESIS: Learning User Simulators as Training Environments for Interactive Agents
Oct 7, 2026Hoang Phan, Dat Huynh, Andrey Zhmoginov +7User Simulation LLM Agent Training
Meta Superintelligence Labs · New York University · University of Wisconsin - Madison
On KL-Regularized Policy Optimization
Oct 6, 2026Yifan ZhangRL for LLM Agents KL-Regularized RL
Princeton University
GraphOPD: Graph-Augmented On-Policy Distillation for LLM Agents
Oct 6, 2026Bohan Lin, Liyi Chen, Zhuoning Guo +5LLM Agent Training On-Policy Distillation
University of Science and Technology of China · Xiaohongshu · The Hong Kong University of Science and Technology (HKUST-GZ)
VETTA: Coordinating Turn- and Token-Level Credit Assignment for Multi-Turn LLM Agents
Oct 6, 2026Jiaju Chen, Min Yang, Jinghua Piao +4Token-Level Credit Assignment Credit Assignment in RL
University of Science and Technology of China · Zhongguancun Academy · Shandong University +2
RELACE: retrospective likelihood-based action credit estimation for long-horizon language agents
Oct 5, 2026Sayak Chakrabarti, Sathish Reddy IndurthiGroup Relative Policy Optimization Credit Assignment in RL
Columbia University · Zoom Communications, Inc.
MemPilot: Orchestrating On-Demand Multimodal Memory Curation for LLM Agents
Oct 5, 2026Haozhen Zhang, Haodong Yue, Quanyu Long +6LLM Agent Memory LLM Memory
Nanyang Technological University · Tsinghua University · University of Illinois Urbana-Champaign
CLIFT: Conformal Self-Verification for Web Agent Training and Test-Time Scaling
Oct 5, 2026Yifan Zhang, Yutong Dai, Viraj Prabhu +3Test-Time Scaling RL for LLM Agents
Salesforce AI Research
AMBER: Training Long-Horizon Web Agents through Append-Only Memory
Oct 5, 2026Chinmay Savadikar, Zhaoyu Zhang, Mingyu Zhao +4Long-Horizon LLM Agents Agent Memory Management
North Carolina State University · Shopify
Hierarchical Reinforcement Learning with Stable Temporal Abstraction for Language Model Agents
Oct 4, 2026Shayan Mohajer Hamidi, Yize Cheng, Yuanda Xu +2Reinforcement Learning Hierarchical RL
LinkedIn Corporation · University of Maryland
Sibyl: An Efficient Small-large Model Collaboration Framework for Long-horizon Tasks
Oct 4, 2026Zhewei Fang, Yuxin Zhang, Zhenwei Shao +8LLM Agent Training Long-Horizon LLM Agents
Hangzhou Dianzi University · Fudan University · The Hong Kong University of Science and Technology +3
Long-MDR: Long-Context Reinforcement Learning for Multimodal Deep-Research Agents
Oct 4, 2026On Tai TangReinforcement Learning Deep Research Agents
The Hong Kong University of Science and Technology (HKUST)
Beyond Instruction Following: Learning Grounded Skill-Following with Skill Contracts
Oct 4, 2026Jianghan Shen, Zhenjie Liu, Yue Li +10Reinforcement Learning LLM Agent Skill Learning
Nanjing University · Shanghai Artificial Intelligence Laboratory · Peking University +3
Do Your Own Research: Learning to Forecast by Learning to Search
Oct 1, 2026Yusuf Afifi, Artur Kiulian, Anton Polishko +3Agentic RL Forecasting Benchmarks
Future Principle
My FAULT: Self-Diagnosis as Credit Assignment in Self-Evolving Agentic Reinforcement Learning
Oct 1, 2026Yihua Zhu, Qianying Liu, Weixu Qiao +10Agentic RL Credit Assignment in RL
Alibaba Group · Kyoto University · NII LLMC +6
RISED: RubrIcs for agentic multi-environment Selection and sElf-Distillation
Oct 1, 2026Jingtan Wang, Sirajul Salekin, Young mok Jung +3Reinforcement Learning Rubric-Based RL
National University of Singapore · Work done during an internship at Apple · Apple
SHARPO: Segment-Level Credit Assignment for Agentic Reinforcement Learning
Sep 30, 2026Xinchen Du, Zhengze Zhou, Wenhui Zhu +4Agentic RL Group Relative Policy Optimization
LinkedIn Corporation · Georgia Institute of Technology
ComputerSD: Online Self-Distillation from Real-Time Feedback for Computer-Use Agents
Sep 30, 2026Yong Du, Tongbo Chen, Zhengxi Lu +5On-Policy Self-Distillation Computer-Use Agents
Zhejiang University · Ant Group
PhantomEnvironments: Training LLM Agents in Fictional Worlds
Sep 30, 2026Anmol Kabra, Swathi Saravana Selvam, Albert Gong +5Multi-Hop QA Reinforcement Learning
Cornell University · Stanford University
From Given to Gathered Evidence: Agentic Learning for Longitudinal Medical Reasoning
Sep 30, 2026Minye Shao, Chaohui Yu, Yixuan Wu +3AI Agent Benchmarks Medical VLMs
Durham University · DAMO Academy, Alibaba Group · Hupan Laboratory +1

Oct 8, 2026, cs.CLJ/K move · Enter open · S save

HarnessSQL: Harness-Native Training for SQL Agents in Realistic Database Environments

Haolin Yang, Jipeng Zhang, Jian Xie, Shuaishuai Gong+2

Hong Kong University of Science and Technology · Microsoft Research · Tsinghua University+1

Text-to-SQL LLM Agent Training LLM Post-Training LLM Agent Harnesses RL for LLM Agents

Open paper AskPDF arXiv

Text-to-SQL models are commonly trained to map questions directly to static queries, whereas real-world database agents operate through stateful, multi-turn interaction with live databases -- inspecting schemas, executing probe queries, diagnosing errors, and revising hypotheses. This creates a critical train-deploy mismatch, as the execution harness that mediates this interaction is introduced only at inference time. To bridge this gap, we propose HarnessSQL, a harness-native post-training framework that preserves the full interaction structure throughout both supervised fine-tuning and reinforcement learning. HarnessSQL builds isolated, executable database environments paired with hidden execution oracles, rolls out teachers directly inside the target SQL harness, and retains only verified trajectories for full-sequence SFT, followed by execution-reward RL. Across Spider 2.0-SQLite, HarnessSQL dramatically boosts the execution accuracy of compact models, raising Qwen3-8B from 15.5% to 45.2% and Qwen3-14B from 22.2% to 54.8%, while transferring effectively to out-of-distribution interactive benchmarks such as BIRD-Interact and LiveSQLBench. Our findings demonstrate that training database agents directly within their execution harness is essential for mastering complex, long-horizon database workflows.

Momentum

Related topics

Latest papers 402