**Why Multi-Step Tool-Use Reinforcement Learning Collapses and How Supervisory S

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

> Tool use enables large language models (LLMs) to perform complex tasks, and recent agentic reinforcement learning (RL) methods show promise for enhancing model capabilities. However, RL alone often leads to instability or limited gains in tool-use tasks. In our experiments, some models exhibit catastrophic collapse, where performance abruptly drops and tool-invocation structures fail. The analysis reveals that these...

**来源信息**
- **来源**：HuggingFace Daily Papers（社区热门论文）
- **分类**：论文
- **发布时间**：2026-06-26 08:00（北京时间）
- **原文**：[打开原文](https://huggingface.co/papers/2606.26027)