ResearchTasksMalicious Action DetectionFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast Updatedstress test 100-seed redcode_style profilePrior exploration baseline100AMR4May 19, 2026100-seed stress orig profile (test)Prior exploration baseline100AMR4May 19, 2026tau-Bench retail (test)GPT-OSS-Safeguard 20B7.8TPR4Mar 18, 2026WebArenaGPT-OSS-Safeguard 20B7.46TPR4Mar 18, 2026