← back to paper

Review history

arxiv: 2607.10611 · 2 revisions

M+Adam: Low-Precision Training via Additive-Multiplicative Optimization

1. 2026-08-02 CONDITIONAL MODERATE v1.3.0-alltime-deepseek novelty 6.0
   171939 ms 55767 in 17662 out 2026-08-02T07:11:25.171258+00:00
2. 2026-07-14 CONDITIONAL HIGH v1.1.0-grok45 novelty 5.0
   33505 ms 61470 in 3116 out 2026-07-14T10:28:19.169529+00:00