Hamish Ivison

The Delta Learning Hypothesis: Preference Tuning on Weak Data can Yield Strong Gains

Tulu 3: Pushing Frontiers in Open Language Model Post-Training

2 OLMo 2 Furious (COLM’s Version)

Personalizing Reinforcement Learning from Human Feedback with Variational Preference Learning

Unpacking DPO and PPO: Disentangling Best Practices for Learning from Preference Feedback