Stable Preference Optimization for LLMs: A Bilevel Approach Beyond Direct Preference Optimization
This addresses the problem of unstable and biased alignment in large language models for AI researchers and practitioners, offering an incremental improvement over DPO.
The paper identifies theoretical limitations in Direct Preference Optimization (DPO) for aligning language models with human preferences, showing it is sensitive to initialization and can misallocate probability mass, then proposes a bilevel optimization framework called stable preference optimization that improves reasoning accuracy and alignment on benchmarks.
Direct Preference Optimization (DPO) has emerged as a popular and efficient alternative to reward modeling and reinforcement learning for aligning language models with human preferences. Despite its empirical success, the theoretical properties and intrinsic limitations of DPO remain underexplored. In this work, we first present a comprehensive analysis of DPO's dynamics from a probability evolution perspective. Our analysis reveals that DPO is highly sensitive to initialization. It also tends to misallocate probability mass, which can inadvertently shift probability toward irrelevant or undesired responses. This misallocation may unintentionally reinforce model bias, thereby compromising both the stability of model alignment and the consistency with intended preferences. Motivated by these theoretical findings, we propose a theoretically grounded bilevel optimization framework that tightly integrate supervised fine-tuning with an enhanced DPO objective a.k.a. stable preference optimization. Our approach introduces a principled regularization scheme to explicitly encourage absolute probability improvement for preferred outputs, while maintaining stable optimization dynamics. Experiments on challenging reasoning and summarization benchmarks elucidate that our method consistently improves reasoning accuracy and better aligns output distributions with intended preferences, outperforming standard DPO. Stable preference optimization provides new insights into the design of preference-based alignment objectives and opens up new avenues towards more reliable and interpretable language model alignment.