How Far Can In-Context Alignment Go? Exploring the State of In-Context Alignment

Heyan Huang, Yinghao Li, Huashan Sun, Yu Bai, Yang Gao


Abstract
Recent studies have demonstrated that In-Context Learning (ICL), through the use of specific demonstrations, can align Large Language Models (LLMs) with human preferences known as In-Context Alignment (ICA), indicating that models can comprehend human instructions without requiring parameter adjustments. However, the exploration of the mechanism and applicability of ICA remains limited. In this paper, we begin by dividing the context text used in ICA into three categories: format, system prompt, and example. Through ablation experiments, we investigate the effectiveness of each part in enabling ICA to function effectively. We then examine how variants in these parts impact the model’s alignment performance. Our findings indicate that the example part is crucial for enhancing the model’s alignment capabilities, with changes in examples significantly affecting alignment performance. We also conduct a comprehensive evaluation of ICA’s zero-shot capabilities in various alignment tasks. The results indicate that compared to parameter fine-tuning methods, ICA demonstrates superior performance in knowledge-based tasks and tool-use tasks. However, it still exhibits certain limitations in areas such as multi-turn dialogues and instruction following. Source codes and scripts are available at https://github.com/li-aolong/how-far-can-ica-go.
Anthology ID:
2024.findings-emnlp.504
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:
8623–8644
Language:
URL:
https://aclanthology.org/2024.findings-emnlp.504
DOI:
10.18653/v1/2024.findings-emnlp.504
Bibkey:
Cite (ACL):
Heyan Huang, Yinghao Li, Huashan Sun, Yu Bai, and Yang Gao. 2024. How Far Can In-Context Alignment Go? Exploring the State of In-Context Alignment. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 8623–8644, Miami, Florida, USA. Association for Computational Linguistics.
Cite (Informal):
How Far Can In-Context Alignment Go? Exploring the State of In-Context Alignment (Huang et al., Findings 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.findings-emnlp.504.pdf