CLAIMay 27

Integrated and Cross-Architecture Interpretation of LLM Reasoning

arXiv:2605.2800658.9h-index: 8
AI Analysis

For researchers studying LLM reasoning, this work provides a more comprehensive interpretability method by integrating multiple probes, but it is incremental as it combines existing techniques.

The paper introduces an Integrated, cross-Architecture Reasoning (IAR) framework to unify LLM reasoning interpretability by combining Mutual Information Peak (MIP) and Deep-Thinking Ratio (DTR) probes. Experiments on three models across four domains show generalizable interpretation capabilities.

Understanding how LLMs reason is hindered by a practical asymmetry: while their generated outputs are observable, the underlying reasoning patterns remain opaque. Relying on single probes, such as Mutual Information Peak (MIP) or Deep-Thinking Ratio (DTR), risks underestimating the genuine inferential structure. To response this deficiency, we present an Integrated, cross-Architecture Reasoning (IAR) framework, designed to provide a unified approach to LLM reasoning interpretability. Specifically, we first propose to use bandwidth-calibrated MIP coupled with Tukey IQR peak-detection to isolate reasoning-crucial tokens at the output layer. Second, we performed an overlap analysis between MIP-picked tokens and DTR-deep tokens to trace the cross-layer trajectories of those tokens. This also discloses whether reasoning-crucial tokens are computation-intensive as well, further facilitating to understand how reasoning patterns evolve across model layers. Finally, we apply a Jaccard stability metric over multi-domain problems to verify if the MIP-identified tokens are reasoning quality-guaranteed. Extensive experiments on three models (Qwen-7B, Qwen-14B, and Llama-8B) across four domains (mathematics, code, logic, and common sense) demonstrate IAR's generalizable interpretation capabilities across architectures.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes