Deeper Insights Without Updates: The Power of In-Context Learning Over Fine-Tuning

Qingyu Yin, Xuzheng He, Chak Tou Leong, Fan Wang, Yanzhao Yan, Xiaoyu Shen, Qiang Zhang


Abstract
Fine-tuning and in-context learning (ICL) are two prevalent methods in imbuing large language models with task-specific knowledge. It is commonly believed that fine-tuning can surpass ICL given sufficient training samples as it allows the model to adjust its internal parameters based on the data. However, this paper presents a counterintuitive finding: For tasks with implicit patterns, ICL captures these patterns significantly better than fine-tuning. We developed several datasets featuring implicit patterns, such as sequences determining answers through parity or identifying reducible terms in calculations. We then evaluated the models’ understanding of these patterns under both fine-tuning and ICL across models ranging from 0.5B to 7B parameters. The results indicate that models employing ICL can quickly grasp deep patterns and significantly improve accuracy. In contrast, fine-tuning, despite utilizing thousands of times more training samples than ICL, achieved only limited improvements. We also proposed circuit shift theory from a mechanistic interpretability’s view to explain why ICL wins.
Anthology ID:
2024.findings-emnlp.239
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2024
Month:
November
Year:
2024
Address:
Miami, Florida, USA
Editors:
Yaser Al-Onaizan, Mohit Bansal, Yun-Nung Chen
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
4138–4151
Language:
URL:
https://aclanthology.org/2024.findings-emnlp.239/
DOI:
10.18653/v1/2024.findings-emnlp.239
Bibkey:
Cite (ACL):
Qingyu Yin, Xuzheng He, Chak Tou Leong, Fan Wang, Yanzhao Yan, Xiaoyu Shen, and Qiang Zhang. 2024. Deeper Insights Without Updates: The Power of In-Context Learning Over Fine-Tuning. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 4138–4151, Miami, Florida, USA. Association for Computational Linguistics.
Cite (Informal):
Deeper Insights Without Updates: The Power of In-Context Learning Over Fine-Tuning (Yin et al., Findings 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.findings-emnlp.239.pdf