conftrace_

reference policy

6 papers

Explore in graph

Co-occurring keywords · papers mentioning both

kl divergence (2) direct preference optimization (2) occupancy measure (1) multi-agent system (1) game-theoretic planning (1) offline alignment (1) iterative expectation (1) win rate (1) proxy reward function (1) alignment (1)

Papers

Correlated Proxies: A New Definition and Improved Mitigation for Reward Hacking ICLR 2025 Learn Your Reference Model for Real Good Alignment ICLR 2025 Theoretical guarantees on the best-of-n alignment policy ICML 2025 Understanding Reference Policies in Direct Preference Optimization NAACL 2025 Blending Data-Driven Priors in Dynamic Games RSS 2024 Reference-Based POMDPs NeurIPS 2023