ResearchTasksAction-component payoff optimizationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedL3 warmup (off-diagonal)Reciprocity Gradient4.17Per-Interaction Payoff8May 12, 2026HybridCoop+AllD (off-diagonal)TD32.05Per-interaction Payoff6May 12, 2026L6 warmup discriminative cellReciprocity Gradient3.94Per-interaction Payoff4May 12, 2026