一作ACL 2026Findings· Findings of ACL
PRISM - MCTS : Learning from Reasoning Trajectories with Metacognitive Reflection
Siyuan Cheng · Bozhong Tian · Yanchao Hao · Zheng Wei
摘要
The emergence of reasoning models, exemplified by OpenAI o1, signifies a transition from intuitive to deliberative cognition, effectively reorienting the scaling laws from pre-training paradigms toward test-time computation. While Monte Carlo Tree Search (MCTS) has shown promise in this domain, e...