作者位次 #3/6ACL 2026Poster· ACL Industry Track
Thinking with Reasoning Skills: Fewer Tokens, More Accuracy
Guangxiang Zhao · Qilong Shi · Xusen Xiao · Xiangzheng Zhang · Tong Yang · Lin Sun
摘要
Reasoning LLMs often spend substantial tokens on long intermediate reasoning traces (e.g., chain-of-thought) when solving new problems. We propose to summarize and store reusable reasoning skills distilled from extensive deliberation and trial-and-error exploration, and to retrieve these skills a...