Rahul Jain

Full Professor, University of Southern California

1 paper at NeurIPS 2025

Homepage· OpenReview· Semantic Scholar· Google Scholar

Poster Session 3

1 paper

Thursday, December 4, 2025 · 11:00 AM → 2:00 PM

Exhibit Hall C,D,E

Robust LLM Alignment via Distributionally Robust Direct Preference Optimization

#3910 · Zaiyan Xu, Sushil Vemuri, Kishan Panaganti, Dileep Kalathil, Rahul Jain, Deepak Ramachandran

We address distributional shift among diverse preferences with robust DPO: Wasserstein DPO (WDPO) and Kullback–Leibler DPO (KLDPO). Finite-sample guarantees, tractable gradient-based algorithms for hard DRO objectives, strong empirical robustness.