In this paper, we aim to align large language models with the ever-changing, complex, and diverse human values (e.g., social norms) across time and locations. This presents a challenge to existing alignment techniques, such as supervised fine-tuning, which internalize values within model parameters. To overcome this, we propose an On-the-fly Preference Optimization (OPO) method, which is a real-time alignment that works in a streaming way. It employs an external memory to store established rules for alignment, which can constrain LLMs' behaviors without further training, allowing for convenient updates and customization of human values. We also introduce a scalable evaluation to assess the proposed method more effectively. Experimental results on both human-annotated and auto-generated questions from legal and moral domains indicate the effectiveness of the proposed OPO method. Our code and data are released at https://github.com/GAIR-NLP/OPO.
@article{arxiv.2312.15907,
title = {Align on the Fly: Adapting Chatbot Behavior to Established Norms},
author = {Chunpu Xu and Steffi Chern and Ethan Chern and Ge Zhang and Zekun Wang and Ruibo Liu and Jing Li and Jie Fu and Pengfei Liu},
journal= {arXiv preprint arXiv:2312.15907},
year = {2023}
}