Pre-DPO: Improving Data Utilization in Direct Preference Optimization Using a Guiding Reference Model
Paper
•
2504.15843
•
Published
•
18
Junjie Lu
Lux0926
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
about 1 month ago
Exploring Data Scaling Trends and Effects in Reinforcement Learning from
Human Feedback
authored
a paper
2 months ago
ML-Bench: Large Language Models Leverage Open-source Libraries for
Machine Learning Tasks
Organizations
