arXiv Machine Learning June 25, 2026 By Yupu Hao, Zhuoran Jin, Huanxuan Liao, Kang Liu, Jun Zhao

Why Multi-Step Tool-Use Reinforcement Learning Collapses and How Supervisory Signals Fix It

Read the original on arXiv Machine Learning →

arXiv:2606. 26027v1 Announce Type: cross Abstract: Tool use enables large language models (LLMs) to perform complex tasks, and recent agentic reinforcement learning (RL) methods show promise for enhancing model capabilities.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

arXiv AI

4d ago

ROSS: Relearning from Self-Generated Rollouts through Selective Supervision

arXiv:2609.35954v1 Announce Type: cross Abstract: Large language model post-training generates self-generated rollouts through reinforcement learning and on-policy distillation, yet this experience i...

By Zhiwei Zhang, Huayu Deng, Fei Zhao, Jiayan Fu, Bin Liang, Kam-Fai Wong, Mu Chuan

llms agents reinforcement-learning fine-tuning efficiency benchmarks

arXiv AI

Aug 19

SOD: Step-wise On-policy Distillation for Small Language Model Agents

SOD: Step-wise On-policy Distillation for Small Language Model Agents proposes a new framework that adaptively reweights distillation strength at each reasoning step based on step-level divergence. This approach mitigates cascading errors in tool-integrated reasoning by attenuating misleading teacher signals in high-divergence regions while preserving dense guidance where student and teacher align. Experiments on math, science, and code benchmarks show up to 20.86% improvement over the second-best baseline, with a 0.6B student scoring 26.13% on AIME 2025.

By Qiyong Zhong, Mao Zheng, Mingyang Song, Xin Lin, Jie Sun, Houcheng Jiang, Xiang Wang, Junfeng Fang

llms agents reinforcement-learning efficiency benchmarks

arXiv AI

Jul 8

Learning to Control LLM Agent Harnesses with Offline Reinforcement Learning

arXiv:2607. 05458v1 Announce Type: cross Abstract: Large language model (LLM) agents are usually improved by changing prompts, models, or hand-written workflows, while the execution harness around the model is treated as fixed infrastructure.

By Haiwen Yi, Xinyuan Song

llms agents reinforcement-learning benchmarks

arXiv Machine Learning

Jul 31

TAPO: Transition-Aware Policy Optimization for LLM Agents

arXiv:2607. 27973v1 Announce Type: new Abstract: Recently, Reinforcement Learning (RL) has emerged as a crucial paradigm for the post-training of Large Language Model (LLM) agents.

By Cong Li, Peixi Peng, Yisen Zhao, Xinyu Hu, Shudong Liu, Zhan Su, Zhuojian Li

llms agents reinforcement-learning

arXiv AI

Jun 30

AI Training Manager: Bounded Closed-Loop Control of Adaptive Training Recipes

arXiv:2606. 29871v1 Announce Type: new Abstract: We present the AI Training Manager, a bounded LLM-based supervisory controller for adaptive machine learning training.

By Anjali Rao, Nikhil Kamalkumar Advani

llms agents reinforcement-learning robotics

Hugging Face Trending Papers

Jul 5

Learning to Control LLM Agent Harnesses with Offline Reinforcement Learning

Large language model (LLM) agents are usually improved by changing prompts, models, or hand-written workflows, while the execution harness around the model is treated as fixed infrastructure. We argue that this harness is itself a learnable control layer.

llms agents reinforcement-learning benchmarks

One email a morning, machine-written

Related stories

ROSS: Relearning from Self-Generated Rollouts through Selective Supervision

SOD: Step-wise On-policy Distillation for Small Language Model Agents

Learning to Control LLM Agent Harnesses with Offline Reinforcement Learning

TAPO: Transition-Aware Policy Optimization for LLM Agents

AI Training Manager: Bounded Closed-Loop Control of Adaptive Training Recipes

Learning to Control LLM Agent Harnesses with Offline Reinforcement Learning