🛰️ Daily AI Frontier
‹ back to 2026-09-01

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

Research Efficiency & Systems

Ranking

Overall 81
Content 100
Popularity 37

Observed public metrics from 1 member.

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

Merged summary

TL;DR - TrainSDC is a low-overhead framework for protecting LLM training from silent data corruption. It targets distinct fault-propagation patterns in Transformer forward and backward passes, keeping training close to fault-free behavior with 1.65%–6.76% runtime overhead.

  • Forward-pass faults are location-dependent, with Q/K-path corruption causing persistent training deviations.
  • Backward-pass vulnerability depends mainly on gradient exponent distributions rather than fault location.
  • TrainSDC combines Q/K-path recomputation, residual-gain monitoring, and exponent-aware gradient scaling.
  • Tests on Llama 3.2-1B and Qwen3-0.6B show resilience under sparse and dense fault injection.

Sources (1)

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

arXiv cs.LG Zhipeng Xia, Haotian Xu, Siyu Yun, Liqi Lin, Hu Liu, Yu Li, Cheng Zhuo 2026-08-31 arXiv:2608.30769
Public signals Semantic Scholar citations 0 · Semantic Scholar influential citations 0
Providers: Hugging Face · N/A OpenAlex · N/A Publisher · N/A Semantic Scholar · Citations 0 · Influential citations 0 X · N/A Fetched 2026-09-25 14:25:21.097304 UTC

TL;DR - TrainSDC is a low-overhead framework for protecting LLM training from silent data corruption. It targets distinct fault-propagation patterns in Transformer forward and backward passes, keeping training close to fault-free behavior with 1.65%–6.76% runtime overhead.

  • Forward-pass faults are location-dependent, with Q/K-path corruption causing persistent training deviations.
  • Backward-pass vulnerability depends mainly on gradient exponent distributions rather than fault location.
  • TrainSDC combines Q/K-path recomputation, residual-gain monitoring, and exponent-aware gradient scaling.
  • Tests on Llama 3.2-1B and Qwen3-0.6B show resilience under sparse and dense fault injection.
item →