List your company

Research

Benchmarks

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

309,131 benchmarksPage 833 of 12,366

Benchmarks

TaskDatasetDomainTrendResultsLast update
LLM Safety EvaluationHARMAMP (malicious instances)adversarial safety securityllm chat instruction30Jun 2, 2026
IMU-based state estimationEuRoC MAV3d vision geometryrobotics embodied ai30Jun 2, 2026
Graph MatchingSynthetic 2D point cloudsgraph30Jun 2, 2026
RegressionCalifornia datasettabular time series30Jun 2, 2026
Graph AlignmentRegular Random Graphs degree 10graph30Jun 2, 2026
Circuit LocalizationIOIcomputer visionbio medical30Jun 2, 2026
Knowledge EditingRippleEdits POPULAR (full requested-edit set)nlpllm chat instruction30Jun 2, 2026
Video CompressionJVET CTC Cb Componentvideo generation editing30Jun 2, 2026
Video CompressionJVET CTC Cr component VTM-11.0 Anchor (All Classes)video generation editing30Jun 2, 2026
Novel View SynthesisTAT Dataset (test)3d vision geometry30Jun 2, 2026
Fact verificationFEVERnlp30Jun 2, 2026
Action RecognitionHAC various domain shiftscomputer vision30Jun 2, 2026
Meta-Reinforcement LearningHopper-Param (ID)robotics embodied ai30Jun 2, 2026
Image ClassificationTinyImageNetcomputer vision30Jun 2, 2026
LLM InferenceQwen3-8B 2k prompts Decode-heavy workloadllm chat instruction30Jun 2, 2026
Agentic SearchAgentic Search Benchmarks Averageagents tool useretrieval search rag30Jun 2, 2026
Multi-hop Question AnsweringMuSiQuenlpretrieval search rag30Jun 2, 2026
Multi-hop Question Answering2WikiMultiHopQAnlpretrieval search rag30Jun 2, 2026
Fréchet RegressionSimulated distributional outputs under linear relationshipstabular time series30Jun 2, 2026
Concept ErasureI2Pcomputer visionnlp30Jun 2, 2026
Human Activity RecognitionMilan (mixed (pure + cross) windows)computer vision30Jun 2, 2026
RLHF Backdoor AttackAnthropic Helpful Harmless prompts (train test)adversarial safety security30Jun 2, 2026
MLLM UnlearningMLLMU-Bench Retain Set 10% ratiollm chat instructionadversarial safety security30Jun 2, 2026
MLLM UnlearningMLLMU-Bench 10% ratio (test)llm chat instructionadversarial safety security30Jun 2, 2026
MLLM UnlearningMLLMU-Bench forget set, 10% ratiollm chat instructionadversarial safety security30Jun 2, 2026