The Flip Side of RLHF: On-Policy Feedback for Reward Model Self-Supervised Improvement
Xiaobo Wang
Yofuria
AI & ML interests
Reward Modeling, Agent Memory, LLM Alignment
Recent Activity
updated a collection 1 day ago
UAPO updated a collection 1 day ago
UAPO updated a collection 1 day ago
UAPOOrganizations
models 0
None public yet
datasets 12
Yofuria/UltraFeedback-binarized-ms-swift-1024
Viewer • Updated • 38.9k • 59
Yofuria/UltraFeedback-ms-swift-1024
Viewer • Updated • 41k • 148
Yofuria/PoliCon
Viewer • Updated • 8.02k • 334 • 2
Yofuria/Skywork-Reward-Preference-80K-v0.2-ms-swift
Viewer • Updated • 77k • 5
Yofuria/UltraFeedback-ms-swift
Viewer • Updated • 64k • 111
Yofuria/llama3-ultrafeedback-armorm-swapped-40
Viewer • Updated • 61.8k • 6
Yofuria/mistral-instruct-ultrafeedback-swapped-40
Viewer • Updated • 62.7k • 6
Yofuria/mistral-instruct-ultrafeedback_multi_pairs
Viewer • Updated • 62.7k • 12
Yofuria/llama3-ultrafeedback-armorm_multi_pairs
Viewer • Updated • 61.8k • 12
Yofuria/gemma2-ultrafeedback-armorm_multi_pairs
Viewer • Updated • 61.5k • 6 • 1