AI RESEARCH
TSPO: Breaking the Double Homogenization Dilemma in Multi-turn Search Policy Optimization
arXiv CS.AI
•
ArXi:2601.22776v2 Announce Type: replace Multi-turn tool-integrated reasoning enables Large Language Models (LLMs) to solve complex tasks through iterative information retrieval. However, current reinforcement learning (RL) frameworks for search-augmented reasoning predominantly rely on sparse outcome-level rewards, leading to a "Double Homogenization Dilemma." This manifests as (1) Process homogenization, where the thinking, reasoning, and tooling involved in generation are ignored.