Teaching Language Models to Self-Improve by Learning from Language Feedback
This addresses the problem of costly and insufficient human feedback for AI alignment, offering a scalable solution for improving language models, though it builds incrementally on existing feedback-based methods.
The paper tackles the challenge of aligning large language models with human intentions by introducing Self-Refinement Tuning (SRT), a method that uses model-generated language feedback to reduce reliance on costly human annotations, resulting in a win rate increase from 9.6% to 25.8% on the AlpacaEval 2.0 benchmark for a 70B parameter model.
Aligning Large Language Models (LLMs) with human intentions and values is crucial yet challenging. Current methods primarily rely on human preferences, which are costly and insufficient in capturing nuanced feedback expressed in natural language. In this paper, we present Self-Refinement Tuning (SRT), a method that leverages model feedback for alignment, thereby reducing reliance on human annotations. SRT uses a base language model (e.g., Tulu2) to generate initial responses, which are critiqued and refined by a more advanced model (e.g., GPT-4-Turbo). This process enables the base model to self-evaluate and improve its outputs, facilitating continuous learning. SRT further optimizes the model by learning from its self-generated feedback and refinements, creating a feedback loop that promotes model improvement. Our empirical evaluations demonstrate that SRT significantly outperforms strong baselines across diverse tasks and model sizes. When applied to a 70B parameter model, SRT increases the win rate from 9.6\% to 25.8\% on the AlpacaEval 2.0 benchmark, surpassing well-established systems such as GPT-4-0314, Claude 2, and Gemini. Our analysis highlights the crucial role of language feedback in the success of SRT, suggesting potential for further exploration in this direction.