ResearchTasksStructural and Semantic EvaluationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedPII-300k 4096 (evaluation)Magentic-One1Structural Score3May 15, 2026PII-300k 512 (eval)APWA100Structural Score3May 15, 2026PII-300k PII-64 (evaluation)Direct100Structural Score3May 15, 2026