Fine-Tuning Language Models Using Formal Methods Feedback
This addresses the cost and scalability limitations of human feedback for fine-tuning language models in autonomous systems, though it appears incremental relative to existing fine-tuning approaches.
The paper tackles the problem of fine-tuning language models for domain-specific autonomous systems tasks without costly human feedback by using formal verification against specifications. The method improved specification satisfaction rates from 60% to 90% in autonomous driving tasks.
Although pre-trained language models encode generic knowledge beneficial for planning and control, they may fail to generate appropriate control policies for domain-specific tasks. Existing fine-tuning methods use human feedback to address this limitation, however, sourcing human feedback is labor intensive and costly. We present a fully automated approach to fine-tune pre-trained language models for applications in autonomous systems, bridging the gap between generic knowledge and domain-specific requirements while reducing cost. The method synthesizes automaton-based controllers from pre-trained models guided by natural language task descriptions. These controllers are verifiable against independently provided specifications within a world model, which can be abstract or obtained from a high-fidelity simulator. Controllers with high compliance with the desired specifications receive higher ranks, guiding the iterative fine-tuning process. We provide quantitative evidences, primarily in autonomous driving, to demonstrate the method's effectiveness across multiple tasks. The results indicate an improvement in percentage of specifications satisfied by the controller from 60% to 90%.