ResearchBenchmarksMeeting claim evaluation on private_data (test)Follow74.5Accuracygpt-5.472.83673.26873.774.132Apr 23, 2026Evaluation ResultsMethodMethodLinksAccuracyCompletenessCoveragegpt-5.4Meetings=22, Evaluator...Meetings=22, Evaluator runs=442026.0474.586.592gpt-4.1Meetings=22, Evaluator...Meetings=22, Evaluator runs=442026.0472.980.181.6