List your company

Research

Benchmarks

Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.

309,131 benchmarksPage 1890 of 12,366

Benchmarks

TaskDatasetDomainTrendResultsLast update
Text-to-Image SafetyUDimage generation editingadversarial safety security18Feb 26, 2026
Text-to-Image SafetyT2VSafetyBenchimage generation editingadversarial safety security18Feb 26, 2026
Indirect Prompt Injection DefenseTrojanToolsadversarial safety security18Feb 26, 2026
Indirect Prompt Injection DefenseCombined Attacksadversarial safety security18Feb 26, 2026
Indirect Prompt Injection DefenseIgnore Instructionadversarial safety security18Feb 26, 2026
Watermark DecodingCOCO (subset)computer vision18Feb 26, 2026
Image Classification11 datasets average CLIP ResNet-101 (test)computer vision18Feb 26, 2026
Attention heatmap predictionSurgAtt-SZPH (test)bio medicalcomputer vision18Feb 26, 2026
Math reasoningMATH-500 (in-distribution)llm chat instructionnlp18May 21, 2026
End-to-End InferenceLMSys-Chat-1M and ShareGPT tracesllm chat instruction18Feb 26, 2026
Autonomous DrivingNAVSIM navtest v2 (test)robotics embodied aicomputer vision18Mar 17, 2026
3D Object DetectionK-Radar (test)3d vision geometrycomputer vision18Jul 3, 2026
RetrievalClipperretrieval search rag18Feb 27, 2026
General Question AnsweringPopQAnlpretrieval search rag18Feb 26, 2026
General Question AnsweringTriviaQAnlpretrieval search rag18Feb 26, 2026
Multi-image context generationMICON-Benchcomputer visionmultimodal18Feb 26, 2026
Speech RecognitionLRS3 high-resourcespeech audio18Feb 26, 2026
Speech RecognitionLRS3 low-resourcespeech audio18Feb 26, 2026
Binary Classificationelectricitytabular time series18May 14, 2026
Fault DiagnosisDIRG (Total)tabular time series18Feb 26, 2026
Fault DiagnosisDIRG (Defective)tabular time series18Feb 26, 2026
Fault DiagnosisDIRG (Non-Defective)tabular time series18Feb 26, 2026
Fault DiagnosisHIT Bearing Dataset (test)tabular time series18Feb 26, 2026
Face Anti-SpoofingWMCAcomputer vision18Mar 4, 2026
Face Anti-SpoofingSIW-M V2computer vision18Mar 4, 2026