ResearchTasksBinary safety hallucination detectionBinary safety hallucination detectionFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedEndoVis18-VQA Out-of-template (val)QA-SNNE (Ent)98Accuracy50Apr 24, 2026PitVQA (External val)QA-SNNE96Accuracy25Apr 24, 2026EndoVis VQA In-template 18 (val)QA-SNNE98Accuracy25Apr 24, 2026PitVQA Open-ended (External val)Llama3.291SNNE5Apr 24, 2026