ResearchTasksTool-based multi-turn dialogueTool-based multi-turn dialogueFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedToolWOZ (test)gpt-4o-FC-JOSH-SFT0.922Avg Reward16Feb 26, 2026