Fetching the paper…

TRAD: Enhancing LLM Agents with Step-Wise Thought Retrieval and Aligned Decision · Around