{"id":"arxiv/2602.08043","url":"https://arxiv.org/abs/2602.08043","tags":["cs.LG","cs.AI","math.NA"],"score":0,"title":"V-ABFT: Variance-Based Adaptive Threshold for Fault-Tolerant Matrix Multiplication in Mixed-Precision Deep Learning","domain":"cs.LG","source":"arxiv","authors":["Yiheng Gao","Qin Hua","Zizhong Chen"],"reviews":0,"subject":"cs.LG","summary":"Algorithm-Based Fault Tolerance (ABFT) is widely adopted to detect silent data corruptions (SDCs) in matrix multiplication, a cornerstone operation in deep learning systems. However, existing threshold determination methods face critical challenges: analytical bounds are overly conservative, while probabilistic approaches like A-ABFT yield thresholds $160$--$4200\\times$ larger than actual rounding errors. We present V-ABFT, a variance-based adaptive threshold algorithm that achieves tighter error bounds by directly modeling the verification difference. By leveraging statistical variance…","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-02-08","reactions":{}}