Spectral Origins of the Self-Correction Blind Spot in Autoregressive Generation

Jul 14, 2026

A new preprint introduces SPARC, a spectral-algebraic theory that formally explains why autoregressive language models (LLMs) often fail to correct their own errors but can fix identical errors when attributed to external sources. The theory proves that this 'self-correction blind spot' arises when the spectral radius of the error-propagation operator is at least one, and it derives a precise activation threshold for correction markers. Experiments across four model backbones and a visual autoregressive probe validate the theory, with predictions matching observed blind-spot rates within 3.2% RMSE.

Why it matters: This work provides the first formal, quantitative model of the self-correction blind spot in autoregressive generation, offering new insights that could improve the reliability of LLMs and related models.

Full story at: arXiv Machine Learning ↗