Pith. sign in

← back to paper

Review history

arxiv: 2607.12790 · 2 revisions

Who Grades the Grader? Co-Evolving Evaluation Metrics and Skills for Self-Improving LLM Agents

  1. 2026-08-02 CONDITIONAL MODERATE v1.3.0-alltime-deepseek novelty 7.0
    317427 ms 21359 in 32394 out 2026-08-02T06:20:26.822585+00:00
  2. 2026-07-15 CONDITIONAL LOW v1.1.0-grok45 novelty 6.5
    24721 ms 6219 in 3136 out 2026-07-15T03:19:39.537086+00:00