OpenAI / MODEL EVIDENCE
1 benchmarks · 2 observations · 1 sources
1 benchmarks · large dot: GPT-4-0314 / human informal proofs; small dots: closest models where reported
GPT-4-0314 / human informal proofs · FIMO (Formal IMO) · Pass@1 · GPT-4-0314 baseline ↗
0% · GPT-4-0314 / human informal proofs · FIMO (Formal IMO) · Pass@8 · GPT-4-0314 baseline · FIMO: A Challenge Formal Dataset for Automated Theorem Proving ↗
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
1 of 1 shared benchmark comparisons
Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.
Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.
Liu, Chengwu, Shen, Jianhao, Xin, Huajian, Liu, Zhengying, Yuan, Ye, Wang, Haiming, Ju, Wei, Zheng, Chuanyang, Yin, Yichun, Li, Lin, Zhang, Ming, Liu, Qun
GPT-4-0314 / human informal proofs