jinachris/Qwen2.5-7B-PURE-PRM
Text Generation
•
8B
•
Updated
•
13
•
1
PURE
PRM and fine-tuned LLM used in our PURE github repo: https://github.com/CJReinforce/PURE
-
-
jinachris/Qwen2.5-7B-PURE-VR
Text Generation • 8B • Updated • 24 • 1 -
jinachris/Qwen2.5-7B-PURE-PRMVR
Text Generation • 8B • Updated • 17 • 1 -
jinachris/PURE-PRM-7B
Token Classification • 7B • Updated • 179 • 4 -
Stop Summation: Min-Form Credit Assignment Is All Process Reward Model Needs for Reasoning
Paper • 2504.15275 • Published • 2