Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1987 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Oriented Object Detection | SODA-A (test) | 17 | Jun 30, 2026 | ||
| Object Detection | PKU-DAVIS-SOD 1.0 (test) | 17 | Jun 30, 2026 | ||
| Object Detection | DSEC-Detection | 17 | Jun 30, 2026 | ||
| Motivational Interviewing Behavior Evaluation | AnnoMI session-level | 17 | Jun 30, 2026 | ||
| Motivational Interviewing Dialogue Generation | MindfulMI (test) | 17 | Jun 30, 2026 | ||
| Motivational Interviewing Global Scoring | MI Global Score Evaluation | 17 | Jun 30, 2026 | ||
| Sequential Recommendation | Pixel-1M (test) | 17 | Jun 30, 2026 | ||
| FVV reconstruction | N3DV | 17 | Jun 30, 2026 | ||
| Knowledge Graph Question Answering | MetaQA 3-hop 1.0 | 17 | Jun 29, 2026 | ||
| Robot Manipulation | LIBERO 50 demos per task | 17 | Jun 26, 2026 | ||
| Image Editing | GEdit-Bench EN | 17 | Jul 7, 2026 | ||
| Single-label environmental sound classification | ESC-50 | 17 | Jun 25, 2026 | ||
| Multi-label sound event classification | AS 20K | 17 | Jun 25, 2026 | ||
| Factual Knowledge Recall | ECLeKTic 1.0 (test) | 17 | Jun 24, 2026 | ||
| Answer Accuracy | BBox-DocVQA | 17 | Jun 24, 2026 | ||
| Grounding Accuracy | DocVQA BBox | 17 | Jun 24, 2026 | ||
| Distortion Diagnosis | ReMMDBench (500-sample split) | 17 | Jun 24, 2026 | ||
| Veracity Classification | ReMMDBench (500-sample split) | 17 | Jun 24, 2026 | ||
| Multiple Clustering | CIFAR10-MC | 17 | Jun 24, 2026 | ||
| Multiple Clustering | Fruit | 17 | Jun 24, 2026 | ||
| Relevance Classification | Six entity-pair tasks T1-T6 (test) | 17 | Jun 24, 2026 | ||
| Forgery Type Identification | Synthbuster | 17 | Jun 24, 2026 | ||
| Forgery Type Identification | GenImage | 17 | Jun 24, 2026 | ||
| Forgery Type Identification | DSO-1 | 17 | Jun 24, 2026 | ||
| Forgery Type Identification | FKG-50K | 17 | Jun 24, 2026 |