← back to paper

Review history

arxiv: 2607.10481 · 2 revisions

ARMOR: Stabilizing On-Policy LLM RL with Off-Policy Anchor Samples

1. 2026-08-02 CONDITIONAL MODERATE v1.3.0-alltime-deepseek novelty 5.0
   179919 ms 3594 in 4187 out 2026-08-02T07:13:29.236714+00:00
2. 2026-07-14 CONDITIONAL HIGH v1.1.0-grok45 novelty 5.5
   35410 ms 23382 in 3207 out 2026-07-14T11:22:08.240344+00:00