Pith. sign in

← back to paper

Review history

arxiv: 2607.09786 · 3 revisions

Length Penalties Make Chain-of-Thought Less Monitorable

  1. 2026-08-04 CONDITIONAL MODERATE v1.3.0-alltime-deepseek novelty 6.0
    159163 ms 24013 in 16127 out 2026-08-04T04:23:42.170160+00:00
  2. 2026-08-02 CONDITIONAL MODERATE v1.3.0-alltime-deepseek novelty 6.0
    109918 ms 27858 in 11532 out 2026-08-02T08:01:34.075457+00:00
  3. 2026-07-14 CONDITIONAL HIGH v1.1.0-grok45 novelty 6.5
    40138 ms 33807 in 3876 out 2026-07-14T15:44:42.520985+00:00