OmniAI-ZJU/NuminaMath-Cot-Distillation-100K
Viewer • Updated • 102k • 27
None defined yet.
VLA-Corrector: Lightweight Detect-and-Correct Inference for Adaptive Action Horizon
GFT: From Imitation to Reward Fine-Tuning with Unbiased Group Advantages and Dynamic Coefficient Rectification