Pith. sign in

← back to paper

Review history

arxiv: 2607.09796 · 2 revisions

Metadata-Free Meta-Reweighted Direct Preference Optimization under Noisy Preference Labels

  1. 2026-08-02 CONDITIONAL MODERATE v1.3.0-alltime-deepseek novelty 6.0
    213968 ms 30985 in 23333 out 2026-08-02T07:54:27.246641+00:00
  2. 2026-07-14 CONDITIONAL HIGH v1.1.0-grok45 novelty 6.0
    36781 ms 31674 in 4029 out 2026-07-14T15:36:00.475949+00:00