Pith. sign in

← back to paper

Review history

arxiv: 2607.03640 · 2 revisions

Revealing Hidden Model Behaviors with Task-Specific Self-Reports

  1. 2026-08-02 CONDITIONAL MODERATE v1.3.0-alltime-deepseek novelty 6.0
    190305 ms 42912 in 17503 out 2026-08-02T08:47:24.261049+00:00
  2. 2026-07-12 CONDITIONAL HIGH v1.1.0-grok45 novelty 6.0
    28930 ms 43827 in 3309 out 2026-07-12T00:59:47.134255+00:00