AsyncFlow combines a distributed streaming data queue with delayed parameter updates to improve RL post-training throughput by 1.59x on average over verl on Ascend clusters.
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
-
AsyncFlow: An Asynchronous Streaming RL Framework for Efficient LLM Post-Training
AsyncFlow combines a distributed streaming data queue with delayed parameter updates to improve RL post-training throughput by 1.59x on average over verl on Ascend clusters.