Control Barrier Function for Aligning Large Language Models
Abstract
This article 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.
Fields of Study & Subjects
Digital Rights Management and Security
Reservoir Engineering and Simulation Methods
Tunneling and Rock Mechanics
Publication details
- Journal / source
- IEEE Transactions on Control Systems Technology
- Publisher
- Institute of Electrical and Electronics Engineers (IEEE)
- Volume
- 34
- Issue
- 4
- Pages
- 1821–1830
- Publication date
- 27 March 2026
- Language
- EN