Bootstrapping Language Models with DPO Implicit Rewards
Paper
•
2406.09760
•
Published
•
38
dpo
-
-
BPO: Supercharging Online Preference Learning by Adhering to the Proximity of Behavior LLM
Paper • 2406.12168 • Published • 7 -
WPO: Enhancing RLHF with Weighted Preference Optimization
Paper • 2406.11827 • Published • 14 -
Step-DPO: Step-wise Preference Optimization for Long-chain Reasoning of LLMs
Paper • 2406.18629 • Published • 40 -
Self-Training with Direct Preference Optimization Improves Chain-of-Thought Reasoning
Paper • 2407.18248 • Published • 30 -
Meta-Rewarding Language Models: Self-Improving Alignment with LLM-as-a-Meta-Judge
Paper • 2407.19594 • Published • 19