作者位次 #4/4ACL 2026Poster· Natural Language Processing
To Think or Not to Think: The Hidden Cost of Meta-Training with Excessive C o T Examples
Vignesh Kothapalli · Ata Fatahibaarzi · Hamed Firooz · Maziar Sanjabi
摘要
Chain-of-thought (CoT) prompting combined with few-shot in-context learning (ICL) has unlocked significant reasoning capabilities in large language models (LLMs). However, ICL with CoT examples is ineffective on novel tasks when the pre-training knowledge is insufficient. We study this problem in...