← back to paper

Review history

arxiv: 2607.11263 · 3 revisions

Bringing Back Rule Induction to Fluid Intelligence Research? An Initial Validation of the ARC-AGI Benchmark in Humans

1. 2026-08-02 CONDITIONAL MODERATE v1.3.0-alltime-deepseek novelty 7.0
   97580 ms 22914 in 11266 out 2026-08-02T06:55:54.221811+00:00
2. 2026-07-15 CONDITIONAL HIGH v1.1.0-grok45 novelty 7.0
   27797 ms 28446 in 2492 out 2026-07-15T08:54:06.729251+00:00
3. 2026-07-14 CONDITIONAL MODERATE v1.1.0-grok45 novelty 6.5
   35332 ms 28452 in 2312 out 2026-07-14T05:54:37.816499+00:00