ResearchDatasetsMacro-averagedMacro-averagedFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsLLM security defenseMacro-averaged GPT-4o, Llama-3 8B, and Mistral-7B (test)11.3Attack Success Rate (ASR)6
LLM security defenseMacro-averaged GPT-4o, Llama-3 8B, and Mistral-7B (test)11.3Attack Success Rate (ASR)6