← back to paper

Review history

arxiv: 2507.19017 · 2 revisions

MindSpeed RL: Distributed Dataflow for Scalable and Efficient RL Training on Ascend NPU Cluster

1. 2026-08-15 CONDITIONAL MODERATE v1.4.0-alltime-deepseek-medium novelty 6.0
   135549 ms 12183 in 14417 out 2026-08-15T18:02:46.864196+00:00
2. 2026-08-06 CONDITIONAL MODERATE v1.4.0-alltime-deepseek-medium novelty 6.0
   106287 ms 12183 in 10764 out 2026-08-06T14:27:19.370941+00:00