at2po-agentic-tree-search-optimization

【效率与协作】Optimize multi-turn agent policies via entropy-guided tree expansion and turn-level credit assignment. AT²PO addresses exploration diversity, sparse credit signal, and policy misalignment problems in LLM agents through structured tree search and turn-aware policy updates.

独立 GitHub 快照 · 自动分类 · 以原作者文档为准