Rule-Guided Feedback: Enhancing Reasoning by Enforcing Rule Adherence in Large Language Models
This addresses the challenge of making LLMs more reliable and constrained in their outputs for users in AI and NLP applications, though it appears incremental as it builds on existing feedback paradigms.
The paper tackles the problem of improving Large Language Model (LLM) performance by enforcing rule adherence through a teacher-student feedback framework, resulting in significant enhancements across diverse tasks like puzzles, writing, and reasoning.
In this paper, we introduce Rule-Guided Feedback (RGF), a framework designed to enhance Large Language Model (LLM) performance through structured rule adherence and strategic information seeking. RGF implements a teacher-student paradigm where rule-following is forced through established guidelines. Our framework employs a Teacher model that rigorously evaluates each student output against task-specific rules, providing constructive guidance rather than direct answers when detecting deviations. This iterative feedback loop serves two crucial purposes: maintaining solutions within defined constraints and encouraging proactive information seeking to resolve uncertainties. We evaluate RGF on diverse tasks including Checkmate-in-One puzzles, Sonnet Writing, Penguins-In-a-Table classification, GSM8k, and StrategyQA. Our findings suggest that structured feedback mechanisms can significantly enhance LLMs' performance across various domains.