← back to paper

Review history

arxiv: 2604.25907 · 2 revisions

How Fast Should a Model Commit to Supervision? Training Reasoning Models on the Tsallis Loss Continuum

1. 2026-05-08 UNVERDICTED LOW v0.9.0 novelty 7.0
   44958 ms 5742 in 1626 out 2026-05-08T03:05:41.602221+00:00
2. 2026-05-07 UNVERDICTED LOW v0.9.0 novelty 6.0
   68749 ms 5699 in 1336 out 2026-05-07T16:23:55.006338+00:00