ResearchTasksConversational Numerical Reasoning Question AnsweringFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedConvFinQA v1 (dev)APOLLO0.785Execution Accuracy8Feb 26, 2026ConvFinQA v1 (test)Human Expert Performance89.44Execution Accuracy7Feb 26, 2026