Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 837 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Class-wise Calibration | MNIST | 30 | May 22, 2026 | ||
| Top-label Confidence Calibration | SVHN | 30 | May 22, 2026 | ||
| Image Classification Calibration | PACS Sketch | 30 | May 22, 2026 | ||
| Image Classification Calibration | PACS Cartoon | 30 | May 22, 2026 | ||
| Image Classification Calibration | PACS Art | 30 | May 22, 2026 | ||
| Code Completion | RepoBench-P LongBench | 30 | May 22, 2026 | ||
| Code Completion | LCC LongBench | 30 | May 22, 2026 | ||
| Summarization | SAMSum LongBench | 30 | May 22, 2026 | ||
| Question Answering | TriviaQA LongBench | 30 | May 22, 2026 | ||
| Question Classification | TREC LongBench | 30 | May 22, 2026 | ||
| Time-Series Classification | SLEEP | 30 | May 21, 2026 | ||
| Video Generation | CausVid | 30 | May 21, 2026 | ||
| Long-term Memory | LongMemEval | 30 | Jun 4, 2026 | ||
| Reinforcement Learning from Verifiable Rewards | HEAD-QA | 30 | May 21, 2026 | ||
| Response Type Prediction | Short-Clips (test) | 30 | May 21, 2026 | ||
| Infrared Small-Target Detection | SIRST v1 | 30 | May 21, 2026 | ||
| Multi-turn Dialogue | MT-Bench | 30 | May 20, 2026 | ||
| Text-to-Image Generation | Text-to-Image Generation Dataset | 30 | May 20, 2026 | ||
| Distribution Estimation | Hickling Student-t benchmark original (test) | 30 | May 20, 2026 | ||
| Feature Attribution | ImageNet (test) | 30 | May 20, 2026 | ||
| Attribution | ImageNet 2012 | 30 | May 20, 2026 | ||
| Attribution | Oxford 102 Flower | 30 | May 20, 2026 | ||
| Unconditional Text Generation | OpenWebText (OWT) (test) | 30 | May 20, 2026 | ||
| Attributing Multimodal Foundation Model Errors | ImageNet misclassified samples (val) | 30 | Jun 17, 2026 | ||
| Surface Normal Prediction | NYU v2 (val) | 30 | May 20, 2026 |