# RAGEN-2: Reasoning Collapse in Agentic RL

> A diagnosis and a fix for a failure mode in RL-trained agents: models that look diverse but are actually copy-pasting a template.

- Org: Manling Li et al.
- Date: 2026-04-07
- Category: paper, algorithm
- Source: https://arxiv.org/abs/2604.06268
- Canonical page: https://ai-tldr.dev/releases/ragen-2-reasoning-collapse/

## In short

Identifies 'template collapse' in multi-turn RL-trained agents and proposes an SNR-aware prompt filter that lifts reasoning across planning, math, web nav, and code.

## What it is

RAGEN-2 is the follow-up paper from the original RAGEN team — a group including researchers from Stanford, Northwestern, Microsoft and the University of Washington, led by Manling Li. The paper studies why reinforcement-learning training of multi-turn agents often destabilises, and names a specific failure mode they call 'template collapse.' In template collapse, the model's outputs look diverse by entropy but they no longer respond to different inputs — the agent is effectively regurgitating a template.

## How it works

The authors argue that entropy alone cannot detect template collapse. They decompose reasoning quality into within-input diversity (entropy) and cross-input distinguishability (mutual information) and show that mutual information correlates with downstream performance far better than entropy does. They then explain template collapse via a signal-to-noise-ratio mechanism and propose 'SNR-aware filtering' — selecting prompts for RL training based on reward variance. The fix improves performance across planning, math reasoning, web navigation and code execution.

## Why it matters

Agentic RL is the approach everyone is using to train the next generation of tool-using models, and yet nobody understands exactly when it breaks. RAGEN-2 gives you a concrete diagnostic (mutual information, not entropy) and a concrete fix (SNR-aware filtering) that you can drop into existing RL pipelines.

## Who it's for

Anyone training or debugging RL-finetuned agent models.

## Links

- [Paper](https://arxiv.org/abs/2604.06268)
- [Repo](https://github.com/RAGEN-AI/RAGEN)
- [HF page](https://huggingface.co/papers/2604.06268)