PrototypeCommon
CJReinforce/PURE
Official code for the paper, "Stop Summation: Min-Form Credit Assignment Is All Process Reward Model Needs for Reasoning"
- Owner
- CJReinforce (personal account)
- Stars
- Stars: 174
- Forks
- Forks: 8
- Language
- Python
- License
- None
- Created
- 2025
- Topics
- llmmathematicso1r1reasoningreinforcement-finetuningreinforcement-learningrl
- Tier
- Prototype (100 to 249 ★) What this tier means
More repos by CJReinforce
All 2 repos by CJReinforce- CJReinforce/PikPakAutoOfflineDownloadBotStars: 511PythonAlpha
Nearby in Python
All Python repos- bgts-ai-org/bgts-context-engineStars: 174PythonPrototype
- tinnel123666888/OmniJevStars: 174PythonPrototype
- try2love/codex-mobile-bridgeStars: 174PythonPrototype
- justquick/django-varnishStars: 174PythonPrototype
- satanas/TurpialStars: 174PythonPrototype
- ccnmtl/fdfgenStars: 174PythonPrototype