Pre-DPO: Improving Data Utilization in Direct Preference Optimization Using a Guiding Reference Model
Paper
•
2504.15843
•
Published
•
11
khazzz1c
khazic
·
AI & ML interests
Reinforcement Learning
Recent Activity
upvoted
a
paper
about 8 hours ago
Pre-DPO: Improving Data Utilization in Direct Preference Optimization
Using a Guiding Reference Model
liked
a model
14 days ago
meta-llama/Llama-3.2-11B-Vision
liked
a dataset
about 2 months ago
opendiffusionai/pexels-photos-janpf
Organizations
