#artificial intelligence
May 2026
Integrated and Cross-Architecture Interpretation of LLM Reasoning
This work proposes to use bandwidth-calibrated MIP coupled with Tukey IQR peak-detection to isolate reasoning-crucial tokens at the output layer, and applies a Jaccard stability metric over multi-domain problems to verify if the MIP-identified tokens are reasoning quality-guaranteed.
Leonardo Matthew Yauw, Wei-Bin Kou, Yujiu Yang
· arXiv.org · 0 citations