Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2464 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Financial Sentiment Analysis | FPB | 15 | Mar 5, 2026 | ||
| Emotion Classification | TweetEval Emotion | 15 | Mar 5, 2026 | ||
| Text Classification | TE-Hate | 15 | Mar 5, 2026 | ||
| Text Classification | TE-Emo | 15 | Mar 5, 2026 | ||
| Text Classification | RT | 15 | Mar 5, 2026 | ||
| Text Classification | SST2 | 15 | Mar 5, 2026 | ||
| Text Classification | FPB | 15 | Mar 5, 2026 | ||
| Temporal Grounding | Charades | 15 | Mar 5, 2026 | ||
| Response Generation Quality | General Response Quality Set | 15 | Mar 5, 2026 | ||
| Question Answering | OpenBookQA | 15 | Mar 5, 2026 | ||
| Safety Evaluation | XSTest Toxic | 15 | Mar 5, 2026 | ||
| Refusal Evaluation | XSTest Seemingly Toxic Subsets | 15 | Mar 5, 2026 | ||
| Ground Truth Alignment and Conversationality | Curated Agricultural Dataset (test) | 15 | Mar 5, 2026 | ||
| Operating System Control | AgentBench OS | 15 | May 8, 2026 | ||
| Document Rectification | DocUNet | 15 | Mar 5, 2026 | ||
| Deepfake Detection | FF++ | 15 | Mar 4, 2026 | ||
| Mathematical Reasoning | AIME 2025 | 15 | Mar 4, 2026 | ||
| Text-to-Image | MS-COCO 5k image-text pairs | 15 | Mar 4, 2026 | ||
| TikZ code generation | DaTikZ V4 (evaluation) | 15 | Mar 4, 2026 | ||
| Motion localization | MeViS | 15 | Mar 4, 2026 | ||
| Part Segmentation | Two-part objects | 15 | Mar 4, 2026 | ||
| Image Classification | SVHN | 15 | Mar 4, 2026 | ||
| Image Classification | GTSRB | 15 | Mar 4, 2026 | ||
| Multimodal Search | SimpleVQA | 15 | Mar 4, 2026 | ||
| Cross-Domain Recommendation | Amazon Book → Music | 15 | Mar 4, 2026 |