ResearchTasksUser Simulation Quality AssessmentFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedTurn-level Human Evaluation Set adversarial (test)UserLM-R1168Win Rate3Feb 26, 2026Session-level Human Evaluation Set adversarial (test)UserLM-R186Win Rate3Feb 26, 2026