---
title: "What Does Multi-Turn Reinforcement Learning Really Improve in LLM Agents?"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.06483v1"
permalink: "https://feed.astralor.com/items/itm_bfd61d726a49eef70b9eb98e"
published: "2026-11-04T13:24:54.000Z"
score: 56
channel: "tech"
tags: ["LLM","AI Agent","多模态","What Does Multi-Turn","Reinforcement Learning Really","Improve","LLM Agents","While"]
---

# What Does Multi-Turn Reinforcement Learning Really Improve in LLM Agents?

## At a glance

While multi-turn agentic reinforcement learning (RL) excels on long-horizon tasks, its performance gains remain highly sensitive to interaction constraints. Evaluating across sampling breadth $K$ and interaction depth $…

## Key points

- While multi-turn agentic reinforcement learning (RL) excels on long-horizon tasks, its performance gains remain highly sensitive to interaction constraints.
- Evaluating across sampling breadth $K$ and interaction depth $T$ via $Pass@(K,T)$, we show that trajectory-level RL advantages depend on sufficient turn budgets, diminishing significantly when budgets tighten.
- Besides, agentic RL does not uniformly elevate single-step decision quality; rather, its strength concentrates in error recovery and long-horizon resilience.

## Full text

While multi-turn agentic reinforcement learning (RL) excels on long-horizon tasks, its performance gains remain highly sensitive to interaction constraints. Evaluating across sampling breadth $K$ and interaction depth $T$ via $Pass@(K,T)$, we show that trajectory-level RL advantages depend on sufficient turn budgets, diminishing significantly when budgets tighten. Besides, agentic RL does not uniformly elevate single-step decision quality; rather, its strength concentrates in error recovery and long-horizon resilience. To mitigate the deployment risks caused by this budget dependence, we provide a preliminary discussion on the benefits and limitations of process supervision signals and introduce BudgetMix RL as an alternative that varies interaction budgets during training. Experiments demonstrate that BudgetMix RL achieves robust performance across diverse execution settings. Project Page: https://rethinking-agentic-rl.github.io/

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_bfd61d726a49eef70b9eb98e