SOURCE-LINKED INTELLIGENCE

TrainSDC: Characterizing and Mitigating Silent Data Corruption in Large Language Model Training

arXiv · AI, language, vision and robotics · article · Aug 31, 2026 · UTC

LLM training is increasingly vulnerable to silent data corruption (SDC), yet existing protection methods largely treat Transformer computations uniformly because their vulnerability remains poorly understood. We present the first systematic characterization of SDC vulnerability across major computation interfaces in both the forward and backward passes of Transformer training. Our analysis reveals two distinct error propagation mechanisms: forward-pass vulnerability is highly location dependent, with faults on the Q/K path producing persistent training deviations, whereas backward-pass vulnera

Read original source ↗ Open in workspace

recordType

paper

region

Global