← back to paper

Review history

arxiv: 2605.19444 · 2 revisions

Detecting and Mitigating the Correct-Answer Extinction Window in Test-Time Reinforcement Learning with Majority Voting

1. 2026-06-30 UNVERDICTED LOW v0.9.1-grok novelty 7.0
   30747 ms 5772 in 1272 out 2026-06-30T17:50:29.532154+00:00
2. 2026-05-20 UNVERDICTED LOW v0.9.0 novelty 6.0
   53646 ms 5801 in 1510 out 2026-05-20T07:16:45.204907+00:00