ResearchTasksCode LeakageFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedCodex GPT-5Skill-Stealing Prompt Generation Framework0Exact Match (EM)12Apr 24, 2026Claude Code MiniMax-M2.7Skill-Stealing Prompt Generation Framework60Exact Match (EM)12Apr 24, 2026