List your company

Research

Benchmarks

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

309,131 benchmarksPage 1685 of 12,366

Benchmarks

TaskDatasetDomainTrendResultsLast update
Image ClassificationMNIST regular i.i.d. (test)computer vision20Feb 26, 2026
Graph ClassificationDD TUDatasetgraph20Feb 26, 2026
Ground-state energy calculationSmall moleculesbio medical20Feb 26, 2026
Formation energy predictionMaterials Project (test)bio medicaltabular time series20Feb 26, 2026
Speech SeparationWHAMR!speech audio20Feb 26, 2026
Image DeblurringRealBlur-J alpha = 20computer vision20Feb 26, 2026
Gaussian DenoisingfastMRI (test)computer vision20Feb 26, 2026
Gaussian DenoisingMRI Dataset (test)computer vision20Feb 26, 2026
3D Hand-Object InteractionHO3D v2 (test)3d vision geometryrobotics embodied ai20Feb 26, 2026
Multimodal Humor ExplanationMemeCap, New Yorker, and YesBut Averagemultimodalnlp20Feb 26, 2026
Multimodal Humor ExplanationYesButmultimodalnlp20Feb 26, 2026
Multimodal Humor ExplanationNew Yorkermultimodalnlp20Feb 26, 2026
Multimodal Humor ExplanationMemeCapmultimodalnlp20Feb 26, 2026
Autonomous Driving PlanningNavsim (test)robotics embodied ai20Jul 1, 2026
Real-world UnderstandingMME-RealWorld ENcomputer visionmultimodal20Feb 27, 2026
Eye Tracking3ET+ CVPR AIS Challenge 2024computer vision20Feb 26, 2026
Instance SegmentationMitoEM Human -> Rat R (test)computer vision20Feb 26, 2026
SegmentationBTCV (100% labels)computer vision20Jun 2, 2026
Pulmonary SegmentationLUNA16 100% labelscomputer visionbio medical20Jun 2, 2026
Pulmonary SegmentationLUNA16 (10% labels)computer visionbio medical20Jun 2, 2026
Machine UnlearningCIFAR-10 50% forget setadversarial safety security20Feb 26, 2026
Machine UnlearningCIFAR-100 (50% forget set)adversarial safety security20Feb 26, 2026
Machine UnlearningMUFAC 10% unlearning (train)adversarial safety security20Feb 26, 2026
Machine UnlearningCIFAR-100 10% unlearning (train)adversarial safety security20Feb 26, 2026
Machine UnlearningTiny-ImageNet (TinyIN) 10% unlearning (train)adversarial safety security20Feb 26, 2026