SOURCE-LINKED INTELLIGENCE

How Perturbations Propagate: A Multi-Level Analysis of Robustness in Large Language Models

arXiv · AI, language, vision and robotics · article · Sep 3, 2026 · UTC

Language models encounter typos, corrupted text, altered words, and disrupted token order, yet robustness is usually evaluated only through output behavior. We study how six naturalistic and synthetic input perturbations propagate through decoder-only language models at three levels: output behavior, hidden-state geometry, and attention-head function. We evaluate behavioral effects across four GPT-2 and two Qwen2.5 checkpoints by analyzing layerwise geometry using centered kernel alignment and intrinsic dimension, and examine attention-head responses in GPT-2. Perturbation types produce distin

Read original source ↗ Open in workspace

recordType

paper

region

Global