Klear-Reasoner: Advancing Reasoning Capability via Gradient-Preserving Clipping Policy Optimization
Paper
•
2508.07629
•
Published
•
42
KlearReasoner-8B
KlearReasoner-8B
-
-
CE-GPPO: Controlling Entropy via Gradient-Preserving Clipping Policy Optimization in Reinforcement Learning
Paper • 2509.20712 • Published • 19 -
Kwai-Klear/Klear-Reasoner-8B
8B • Updated • 16 • 19 -
Kwai-Klear/KlearReasoner-MathSub-30K
Viewer • Updated • 30k • 85 • 3 -
Kwai-Klear/KlearReasoner-CodeSub-15K
Viewer • Updated • 15k • 126 • 5 -
Kwai-Klear/Klear-Reasoner-8B-SFT
8B • Updated • 3 • 2