【第212期】Self-Backtracking：自我回溯

Seventy3 · Apr 30, 2025 · 23m

Seventy3：借助NotebookLM的能力进行论文解读，专注人工智能、大模型、机器人算法方向，让大家跟着AI一起进步。

进群添加小助手微信：seventy3_podcast

备注：小宇宙

今天的主题是：

Step Back to Leap Forward: Self-Backtracking for Boosting Reasoning of Language Models

Summary

This paper introduces Self-Backtracking, a novel technique to enhance the reasoning of large language models (LLMs) by enabling them to internally manage a search process with backtracking capabilities. The method trains LLMs to recognize suboptimal reasoning paths and autonomously backtrack to explore alternatives during both training and inference. This internalization of backtracking aims to…