ResearchTasksAgent Preference ModelingFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedObserved Preferences Full FitPer-Agent Per-Goal (Lower Bound)0.092Modelling Loss8May 25, 2026Observed Preferences No Distractors Fit Distractors LossProposed Model0.154Modelling Loss6May 25, 2026Observed Preferences Single-Stage Fit Two-Stage LossProposed Model0.185Modelling Loss6May 25, 2026Observed Preferences K-Fold Cross-ValidationProposed Model0.153Modelling Loss6May 25, 2026
Observed Preferences No Distractors Fit Distractors LossProposed Model0.154Modelling Loss6May 25, 2026