conftrace_

iterative dpo

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

policy improvement oracle (1) rejection sampling strategy (1) reverse-kl regularized contextual bandit (1) alpacaeval 2.0 (1) llm-as-a-judge (1) self-rewarding (1) information theoretical (1) finite-sample guarantee (1) language model (1) instruction following (1)

Papers

Iterative Preference Learning from Human Feedback: Bridging Theory and Practice for RLHF under KL-constraint ICML 2024 Self-Rewarding Language Models ICML 2024