← back to paper

Review history

arxiv: 2607.01232 · 2 revisions

Is One Layer Enough? Training A Single Transformer Layer Can Match Full-Parameter RL Training

1. 2026-07-03 UNVERDICTED LOW v0.9.1-grok novelty 6.0
   35534 ms 5800 in 1126 out 2026-07-03T21:29:33.096598+00:00
2. 2026-07-02 UNVERDICTED LOW v0.9.1-grok novelty 6.0
   42563 ms 5800 in 1200 out 2026-07-02T14:45:50.878572+00:00