Dataset and reward models for "On the Robustness of Reward Models for Language Model Alignment (ICML 2025)"
rm-robustness
community
AI & ML interests
None defined yet.
datasets
5
rm-robustness/ultrafeedback-valid-4-mutual-ood
Viewer
•
Updated
•
11.1k
•
9
rm-robustness/ultrafeedback-valid-3-response-ood
Viewer
•
Updated
•
51.2k
•
9
rm-robustness/ultrafeedback-valid-2-prompt-ood
Viewer
•
Updated
•
11.1k
•
5
rm-robustness/ultrafeedback-valid-1-in-domain
Viewer
•
Updated
•
51.2k
•
4
rm-robustness/ultrafeedback-train
Viewer
•
Updated
•
51.2k
•
7