Pre-DPO: Improving Data Utilization in Direct Preference Optimization Using a Guiding Reference Model
Paper
•
2504.15843
•
Published
•
11
DDDTYXS
DtYXs
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
about 8 hours ago
Pre-DPO: Improving Data Utilization in Direct Preference Optimization
Using a Guiding Reference Model
published
a dataset
1 day ago
DtYXs/llama3.2-3b-ultrafeedback-armorm-binarized
published
a dataset
1 day ago
DtYXs/qwen2.5-7b-ultrafeedback-armorm-binarized
Organizations
