List your company

Research

Benchmarks

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

309,131 benchmarksPage 1176 of 12,366

Benchmarks

TaskDatasetDomainTrendResultsLast update
Anomaly DetectionEmail Spamtabular time series24May 29, 2026
Long Video UnderstandingLongVideoBenchcomputer vision24May 29, 2026
Egocentric Video UnderstandingEgoSchemacomputer vision24May 29, 2026
Memory Poisoning AttackFinancial Agentadversarial safety security24May 29, 2026
Memory Poisoning AttackMedical Agentadversarial safety security24May 29, 2026
Memory Poisoning AttackPersonal Agentadversarial safety security24May 29, 2026
Membership InferenceLIMRadversarial safety security24May 29, 2026
Membership InferenceEurusadversarial safety security24May 29, 2026
EEG Super-ResolutionSEED-IV 62 ch.bio medicaltabular time series24May 29, 2026
EEG Super-ResolutionSEED 62 ch.bio medicaltabular time series24May 29, 2026
Watermark RemovalCelebA-HQ LoRA, w/o teimage generation editing24May 29, 2026
Image ClassificationCIFAR-10 (2-way sharded split)computer vision24May 29, 2026
Factual QASimpleQAnlpretrieval search rag24May 29, 2026
3D Gaussian Splatting WatermarkingBlender and LLFF (test)3d vision geometryimage generation editing24May 29, 2026
Blockchain Anomaly DetectionAlphaHomora (test)tabular time series24May 29, 2026
Image ClassificationCIFAR-100computer vision24May 29, 2026
Inductive dynamic link predictionUN Votegraph24Jun 4, 2026
Photonic Inverse DesignPCF design tasksmiscellaneous24May 29, 2026
Logical planningCountdown (test)llm chat instructionagents tool use24May 29, 2026
Logical planningSudoku (test)llm chat instructionagents tool use24May 29, 2026
Mathematical ReasoningAIME 25llm chat instructionnlp24May 29, 2026
Mathematical ReasoningMATH 500llm chat instructionnlp24May 29, 2026
CTR PredictionKuaiVideo BARS-tuned (test)tabular time series24May 29, 2026
Instruction Following EvaluationPPE-IFEvalllm chat instruction24May 29, 2026
Transaction orderingSynthetic trade sequences log-normal distribution (1,000 trials per n)tabular time series24May 29, 2026