Process Reward Models that Think -- https://arxiv.org/abs/2504.16828

LAUNCH Lab
Enterprise
university
AI & ML interests
Factuality, reasoning, alignment, LLM applications
Recent Activity
Collections
1
spaces
3
models
3
datasets
9
launch/thinkprm-1K-verification-cots
Viewer
•
Updated
•
1k
•
223
•
4
launch/CLASH
Viewer
•
Updated
•
345
•
95
launch/ManyICLBench
Viewer
•
Updated
•
9.6k
•
217
launch/FactBench
Viewer
•
Updated
•
1k
•
47
•
3
launch/gov_report
Viewer
•
Updated
•
58.4k
•
357
•
6
launch/gov_report_qs
Viewer
•
Updated
•
7.87k
•
173
•
2
launch/open_question_type
Viewer
•
Updated
•
4.96k
•
55
•
4
launch/reddit_qg
Viewer
•
Updated
•
720k
•
144
launch/ampere
Viewer
•
Updated
•
400
•
162