Control Barrier Function for Aligning Large Language Models
This addresses the challenge of ensuring safe and aligned text generation for users of LLMs, though it appears incremental as it builds on existing control and alignment methods.
The paper tackles the problem of aligning large language models (LLMs) to generate user-desirable text by proposing a control-based framework that uses a control barrier function (CBF) as a safety filter, resulting in an add-on system that can be applied without fine-tuning the baseline LLM.
This paper proposes a control-based framework for aligning large language models (LLMs) by leveraging a control barrier function (CBF) to ensure user-desirable text generation. The presented framework applies the CBF safety filter to the predicted token generated from the baseline LLM, to intervene in the generated text. The safety filter includes two significant advantages: this safety filter is an add-on type, allowing it to be used for alignment purposes without fine-tuning the baseline LLM, and if there is an evaluation model regarding the desired alignment, it can be directly applied to the filter design. The overall text-generation system is implemented with open-source language models, aiming to generate positive text.