← back to paper

Review history

arxiv: 2607.04819 · 2 revisions

Layer-Parallel Inference Reduces Encrypted Nonlinear Depth in Transformers

1. 2026-07-14 CONDITIONAL HIGH v1.1.0-grok45 novelty 5.5
   32299 ms 15728 in 3795 out 2026-07-14T16:18:47.181084+00:00
2. 2026-07-11 CONDITIONAL HIGH v1.1.0-grok45 novelty 5.0
   31830 ms 15568 in 3609 out 2026-07-11T13:02:05.935357+00:00