AutoSP automates sequence parallelism and long-context activation checkpointing via compilation, enabling up to 2.7x longer training contexts on NVIDIA hardware with negligible throughput loss.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
AutoSP: Unlocking Long-Context LLM Training Via Compiler-Based Sequence Parallelism
AutoSP automates sequence parallelism and long-context activation checkpointing via compilation, enabling up to 2.7x longer training contexts on NVIDIA hardware with negligible throughput loss.