Pith. sign in

← back to paper

Review history

arxiv: 2608.03294 · 2 revisions

Provably Learning Multi-Head Attention with Queries

  1. 2026-08-10 ACCEPT MODERATE v1.4.0-alltime-deepseek-medium novelty 7.0
    202425 ms 41461 in 19810 out 2026-08-10T04:27:08.022460+00:00
  2. 2026-08-05 ACCEPT MODERATE v1.4.0-daily-deepseek-medium novelty 8.0
    338298 ms 41434 in 30302 out 2026-08-05T21:34:21.485070+00:00