ResearchTasksHuman agreement evaluationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedInteractive Agent Evaluation CriteriaOnline Agent-as-a-Judge0.7Average Agreement Score3Jun 9, 2026PsyCrisis (sampled)PsyCrisis LLM-as-Judge (Ours)0.484Pearson R (Overall)3Feb 26, 2026