← back to paper

Review history

arxiv: 2608.03457 · 2 revisions

LLaDA MoE v2: Scaling Mixture-of-Experts Diffusion Language Models

1. 2026-08-15 CONDITIONAL MODERATE v1.4.0-alltime-deepseek-medium novelty 6.0
   125097 ms 24445 in 15197 out 2026-08-15T14:49:29.569428+00:00
2. 2026-08-05 CONDITIONAL MODERATE v1.4.0-daily-deepseek-medium novelty 6.0
   201801 ms 24440 in 23309 out 2026-08-05T18:49:19.294114+00:00