作者位次 #3/8ACL 2026Findings· Findings of ACL
LLM Inductive Reasoning Through Multi-Agent Enhanced M onte C arlo Tree Search
Xiang Li · Yucheng Zhou · Xiangzhi Wei · Zesheng Shi · Haiyuan Wan · Gong Yifan · Fangming Liu · Jing Li
摘要
Existing methods for enhancing the inductive reasoning of large language models (LLMs) at test-time typically depend on iterative self-refinement of hypotheses, which lacks explicit optimization guidance and effective error correction. This often results in superficial rewording and the accumulat...