Entropy-Gradient Inversion: Moving Toward Internal Mechanism of Large Reasoning Models
Junyao Yang, Chen Qian, Kun Wang, Linfeng Zhang, Quanshi Zhang, Yong Liu†, Dongrui Liu†
- Equal contribution · † Corresponding authors
Entropy-Gradient Inversion identifies a robust negative correlation between token entropy and logit gradients as a geometric signature of reasoning capability. CorR-PO incorporates this signature into reward regularization to stabilize reasoning-model reinforcement learning.