← back to paper

Review history

arxiv: 2606.22158 · 2 revisions

Improving Reasoning in Vision-Language Models via Perception Verified Self-Training

1. 2026-07-01 UNVERDICTED LOW v0.9.1-grok novelty 5.0
   48722 ms 5783 in 1300 out 2026-07-01T06:27:12.880924+00:00
2. 2026-06-26 UNVERDICTED LOW v0.9.1-grok novelty 6.0
   23014 ms 5808 in 1256 out 2026-06-26T12:12:43.480722+00:00