ResearchTasksLLM agent prompt injection defense evaluationFor AI assistantsMarkdownContent indexLLM agent prompt injection defense evaluationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedAgentDojo without attacktool_filter0.49Total Tokens (M)8Mar 27, 2026