List your company

Research

Benchmarks

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

309,131 benchmarksPage 1462 of 12,366

Benchmarks

TaskDatasetDomainTrendResultsLast update
Open-vocabulary semantic segmentationPascal VOC 20 With Backgroundcomputer vision21Mar 25, 2026
Geometry ReasoningGeometry3K (val)3d vision geometrynlp21Jun 4, 2026
ClusteringCifar10tabular time series21Mar 31, 2026
Solving PDEAllen-Cahnmiscellaneous21Jun 1, 2026
Membership Inference AttackFlickr (test)adversarial safety security21Apr 2, 2026
Image RestorationRealPhoto60computer vision21Apr 28, 2026
Graph structure learning50-node Erdős-Rényi graphs p=0.5graph21Mar 24, 2026
Coronary artery segmentationImageCASbio medicalcomputer vision21Mar 24, 2026
Dementia Risk PredictionNYU cohort 3-year horizonbio medicaltabular time series21Mar 24, 2026
Dementia Risk PredictionNYU cohort 2-year horizonbio medicaltabular time series21Mar 24, 2026
Dementia Risk PredictionNYU cohort 1-year horizonbio medicaltabular time series21Mar 24, 2026
Nudity DetectionNSFW Conceptscomputer vision21Mar 24, 2026
RecommendationAmazon Musictabular time series21Apr 16, 2026
Speech Emotion RecognitionFour OOD (test)speech audio21Mar 24, 2026
Infrared small-target unmixingCSIST100Kcomputer vision3d vision geometry21Mar 24, 2026
3D Microstructure Evolution InferencePMC 3D Mesh (inference)3d vision geometrycomputer vision21Mar 24, 2026
Multiple Choice QuestionsTinyMMLUnlp21Mar 24, 2026
LLM-as-a-JudgePreferenceBenchllm chat instruction21Mar 24, 2026
RegressionCH 50% random extra featurestabular time series21Mar 24, 2026
RegressionYE 50% corrupted features noisetabular time series21Mar 24, 2026
ClassificationOT 50% corrupted features (50% noise)computer visionnlp+321Mar 24, 2026
RegressionMI 50% corrupted features noisetabular time series21Mar 24, 2026
ClassificationJA 50% corrupted features (noise)computer visionnlp+321Mar 24, 2026
RegressionHO 50% corrupted features (noise)tabular time series21Mar 24, 2026
ClassificationHI 50% corrupted featurescomputer visionnlp+321Mar 24, 2026