Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2382 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Point Cloud Denoising | PCNet 50K points, 2% Gaussian noise | 15 | May 27, 2026 | ||
| Point Cloud Denoising | PCNet 50K points 1% Gaussian noise | 15 | May 27, 2026 | ||
| Point Cloud Denoising | PCNet 10K points, 3% Gaussian noise | 15 | May 27, 2026 | ||
| Point Cloud Denoising | PCNet 10K points, 2% Gaussian noise | 15 | May 27, 2026 | ||
| Point Cloud Denoising | PCNet 10K points, 1% Gaussian noise | 15 | May 27, 2026 | ||
| Question Answering | MMLU | 15 | May 27, 2026 | ||
| Anomaly Detection | 15 | May 27, 2026 | |||
| Text Summarization | TL;DR | 15 | May 27, 2026 | ||
| Fact Verification | X-Fact Zero-Shot (ZS) | 15 | May 27, 2026 | ||
| Fact Verification | X-Fact Out-of-Domain (OOD) | 15 | May 27, 2026 | ||
| Fact Verification | X-Fact In-Domain (ID) | 15 | May 27, 2026 | ||
| Image Classification | eight-task benchmark CLIP-ViT (minimum over eight per-task cells) | 15 | May 27, 2026 | ||
| Granularity Estimation | GRANOLA-EQ (test) | 15 | May 27, 2026 | ||
| Granularity Ordering | GRANOLA-EQ (test) | 15 | May 27, 2026 | ||
| Interactive Agent Task | WEBSHOP | 15 | May 27, 2026 | ||
| Interactive Agent Task | ScienceWorld | 15 | May 27, 2026 | ||
| Interactive Agent Task | ALFWORLD | 15 | May 27, 2026 | ||
| Long Video Generation | MSVE-Bench | 15 | May 27, 2026 | ||
| Long Video Generation | MovieBench | 15 | May 27, 2026 | ||
| Long Video Generation | ViStory (Cross-Frame) | 15 | May 27, 2026 | ||
| Long Video Generation | ViStory Self-Consistency | 15 | May 27, 2026 | ||
| Long Video Generation | StoryMem | 15 | May 27, 2026 | ||
| Visual Reasoning | TIR-Bench | 15 | May 27, 2026 | ||
| Image Classification | CIFAR-10 (test) | 15 | Jul 8, 2026 | ||
| Novel View Synthesis | Insta360 scene4 | 15 | May 27, 2026 |