List your company

Agreement with outcome human labels on CUAVerifierBench Internal Dataset

81AccuracyUniversal Verifier (UV)
63.3267.9172.577.09Apr 5, 2026

Evaluation results

MethodLinks
Universal Verifier (UV)Verifier=UV, Backbone=GPT-5.22026.04818164321
Universal VerifierVerifier=UV, Base Model=GPT-5.22026.04818164321
WebJudgeVerifier=WebJudge, Backbone=o4-mini2026.047274443322
WebJudgeVerifier=WebJudge, Base Model=o4-mini2026.047274443322
WebVoy.Verifier=WebVoy., Backbone=GPT-5.22026.047069434410
WebVoy.Verifier=WebVoy., Backbone=GPT-4o2026.046773312445
WebVoyagerVerifier=WebVoy., Base Model=GPT-4o2026.046773312445
WebJudgeVerifier=WebJudge, Backbone=GPT-5.22026.04645833577