ResearchTasksMulti-stage Attack DetectionFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedCSTM-Bench (cross-session shard)Coreset Memory Reader73.1Detection Rate3Apr 24, 2026CSTM-Bench (dilution shard)Per-session judge100Detection Rate3Apr 24, 2026