3 papers
cs.DC2026
Explicit State and Resource Contracts for Low-Precision Pipeline Parallel Training under Captured Graphs
Genlang Chen, Junyi Zhu
CUDA Graphs eliminate launch overheads by replaying tensor operations over static virtual addresses. However, FP8 pipeline training continuously alters the scaling states, microbat…
cs.DC2026
Zero-I/O Fault Recovery for Sharded Deep Learning via Dynamic Framework Dependency Rebinding
Genlang Chen, Junyi Zhu
Distributed model training at scale is frequently interrupted by transient network failures, conventionally forcing cluster managers to abort all processes and roll back to the lat…
cs.DC2026
SemBridge: Compiling Consumer Observations into Cross-Stack Communication Plans
Genlang Chen, Junyi Zhu, Yuanshan Lin
Distributed-tensor systems specify where values reside, while collective systems optimize how requested operations execute. At a boundary between vendor runtimes that cannot share…