LLMs resist low-frequency permanent GPU faults but certain datapaths and precision formats trigger catastrophic training divergence even at moderate fault rates.
Language model evaluation beyond per- plexity,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
method 1
citation-polarity summary
fields
cs.AR 1years
2026 1verdicts
UNVERDICTED 1roles
method 1polarities
use method 1representative citing papers
citing papers explorer
-
LLM-PRISM: Characterizing Silent Data Corruption from Permanent GPU Faults in LLM Training
LLMs resist low-frequency permanent GPU faults but certain datapaths and precision formats trigger catastrophic training divergence even at moderate fault rates.