Sharpness-Aware Minimization in Logit Space Efficiently Enhances Direct Preference Optimization
Direct Preference Optimization (DPO) has emerged as a popular algorithm for aligning pretrained large language models with human preferences, owing to its simplicity and training stability. However, DPO suffers from the recently identified squeezing effect (also known as likelihood displacement), where the probability of preferred responses decreases unintentionally during training. To understand and mitigate this phenomenon, we develop a theoretical framework that models the coordinate-wise dynamics in logit space. Our analysis reveals that negative-gradient updates cause residuals to expand rapidly along high-curvature directions, which underlies the squeezing effect, whereas Sharpness-...
| Authors | Haocheng Luo, Zehang Deng, Thanh-Toan Do et al. |
|---|---|
| Primary category | cs.LG |
| Published | 18 March 2026 |
| Last updated at source | 18 March 2026 |
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.