Intrinsic Mutual Information as a Modulator for Preference Optimization
For practitioners aligning LLMs with human values, RMiPO reduces hyperparameter tuning overhead while improving performance.
RMiPO uses intrinsic response-level mutual information to dynamically modulate preference contributions in offline preference optimization, achieving superior performance while reducing training overhead by over 15%.
Offline preference optimization methods, such as Direct Preference Optimization (DPO), offer significant advantages in aligning Large Language Models (LLMs) with human values. However, achieving optimal performance with these methods typically involves additional hyperparameter tuning, resulting in substantial time overhead. Although prior work has proposed a range of improvements, these methods remain limited in effectiveness and have not fully eliminated reliance on hyperparameter tuning. In this work, we propose RMiPO, a lightweight and efficient framework for offline preference optimization. RMiPO leverages intrinsic Response-level Mutual information for Preference Optimization with hyperparameter modulation, dynamically decoupling preference contributions at negligible additional computational cost. Extensive experimental results demonstrate that RMiPO achieves consistently superior performance over existing methods while reducing training overhead by more than 15\%. Our code is available at https://github.com/liavonpenn/rmipo.