ResearchTasksDafny Code SynthesisFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedDafny Benchmark Overall Vericoding (aggregate)Model union82.2Pass Rate4Jun 1, 2026APPS Vericoding-derived (test)Model union83Pass Rate4Jun 1, 2026Dafny Hard Subset Quality-filtered (val)Multi-turn RLVR31.1Pass Rate2Jun 1, 2026APPS Dafny-derived (val)Initial RLVR53.9Pass Rate1Jun 1, 2026