Chat Vector: A Simple Approach to Equip LLMs with Instruction Following and Model Alignment in New Languages
This provides a simple and efficient solution for adapting LLMs to new languages, addressing data constraints for non-English conversational AI, though it is incremental as it builds on existing model arithmetic techniques.
The authors tackled the problem of enabling instruction following and human value alignment in non-English languages for pre-trained LLMs by introducing a 'chat vector' derived from weight differences between base and chat models, which when added to continual pre-trained models, achieves superior efficacy in instruction following, toxicity mitigation, and multi-turn dialogue without further training.
Recently, the development of open-source large language models (LLMs) has advanced rapidly. Nevertheless, due to data constraints, the capabilities of most open-source LLMs are primarily focused on English. To address this issue, we introduce the concept of $\textit{chat vector}$ to equip pre-trained language models with instruction following and human value alignment via simple model arithmetic. The chat vector is derived by subtracting the weights of a pre-trained base model (e.g. LLaMA2) from those of its corresponding chat model (e.g. LLaMA2-chat). By simply adding the chat vector to a continual pre-trained model's weights, we can endow the model with chat capabilities in new languages without the need for further training. Our empirical studies demonstrate the superior efficacy of the chat vector from three different aspects: instruction following, toxicity mitigation, and multi-turn dialogue. Moreover, to showcase the adaptability of our approach, we extend our experiments to encompass various languages, base models, and chat vectors. The results underscore the chat vector's simplicity, effectiveness, and wide applicability, making it a compelling solution for efficiently enabling conversational capabilities in pre-trained language models. Our code is available at https://github.com/aqweteddy/ChatVector.