← back to paper

Review history

arxiv: 2605.03667 · 2 revisions

ELAS: Efficient Pre-Training of Low-Rank Large Language Models via 2:4 Activation Sparsity

1. 2026-05-07 UNVERDICTED LOW v0.9.0 novelty 5.0
   57901 ms 5581 in 1208 out 2026-05-07T17:01:49.761503+00:00
2. 2026-05-07 UNVERDICTED LOW v0.9.0 novelty 5.0
   25339 ms 5559 in 1149 out 2026-05-07T01:35:37.433920+00:00