Pith. sign in

← back to paper

Review history

arxiv: 2607.12963 · 2 revisions

The Illusion of Robustness: Aggregate Accuracy Hides Prediction Flips under Task-Irrelevant Context

  1. 2026-08-02 ACCEPT MODERATE v1.3.0-alltime-deepseek novelty 6.0
    378089 ms 23972 in 23866 out 2026-08-02T06:13:58.343487+00:00
  2. 2026-07-15 UNVERDICTED LOW v1.1.0-grok45 novelty 5.5
    29134 ms 6077 in 2585 out 2026-07-15T01:57:55.069939+00:00