MMODELYST
Papers/Entropy-Gradient Inversion: Moving Toward Internal Mechanism of Large Reasoning Models
PAP

Entropy-Gradient Inversion: Moving Toward Internal Mechanism of Large Reasoning Models

May 18, 2026

arXiv
Abstract

The advancement of Large Reasoning Models (LRMs) has catalyzed a paradigm shift from reactive ``fast thinking'' text generation to systematic, step-by-step ``slow thinking'' reasoning, unlocking state-of-the-art performance in complex mathematical and logical tasks. However, the field faces \textit{the fundamental gap between token-level behavioral analysis and internal reasoning mechanisms, and the instability of reinforcement learning (RL) for reasoning optimization relying on costly external verifiers}. We identify and formally define \textbf{Entropy-Gradient Inversion}, a robust negative correlation between token entropy and logit gradients that acts as a definitive geometric fingerprint for LRM reasoning capability. Building on this, we propose \textbf{Correlation-Regularized Group Policy Optimization (CorR-PO)}, which embeds this inversion signature into RL reward regularization. Extensive experiments on various reasoning benchmarks across multiple model scales show CorR-PO consistently outperforms state-of-the-art baselines, confirming that stronger inversion directly correlates with superior reasoning performance.

Select text to highlight · click a highlight to remove · saved in this browser only
Authors
Junyao Yang, Chen Qian, Kun Wang, Linfeng Zhang, Quanshi Zhang, Yong Liu, Dongrui Liu
Your notes (browser-local)
saved
arXiv:2605.17770