A slice-level pipeline-parallel schedule with attention-work redistribution that cuts activation memory roughly by the pipeline size and reduces pipeline bubbles for long-context LLM training.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
SlimPipe: Memory-Thrifty and Efficient Pipeline Parallelism for Long-Context LLM Training
A slice-level pipeline-parallel schedule with attention-work redistribution that cuts activation memory roughly by the pipeline size and reduces pipeline bubbles for long-context LLM training.