conftrace_

silent data corruption

4 papers

Explore in graph

Also known as

SDC

Co-occurring keywords · papers mentioning both

llm training (3) hardware fault tolerance (1) large language model (1) online detection (1) bit flip (1) optimizer step (1) hardware failure (1) fault isolation (1) system reliability (1) hardware error (1)

Papers

Safeguarding LLM Training at Scale: Online SDC Detection and Insights from 35 Million GPU Hours OSDI 2026 SDCs in the Wild: Characterizing and Diagnosing SDC-Defective GPUs in Production LLM Training (Operational Systems) OSDI 2026 Understanding Silent Data Corruption in LLM Training ACL 2025 SDC-causing Error Detection Based on Lightweight Vulnerability Prediction ACML 2019