一作ACL 2026Findings· Findings of ACL
Failure makes the agent stronger: Enhancing Accuracy through Structured Reflection for Reliable Tool Interactions
Junhao Su · Yuanliang Wan · Junwei Yang · Hengyu Shi · Tianyang Han · Yurui Qiu · Junfeng Luo
摘要
Tool-augmented large language models (LLMs) are typically trained via supervised imitation learning or coarse-grained reinforcement learning, approaches that primarily optimize one-shot tool calls. Existing practices of self-reflection largely rely on heuristic prompting or unidirectional reasoni...