Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1582 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Mathematical Reasoning | HMMT 26 | 20 | Apr 13, 2026 | ||
| Natural Language Understanding | Wikipedia corpus annotated (train) | 20 | Apr 10, 2026 | ||
| Fact Memorization | Wikipedia corpus annotated (train) | 20 | Apr 10, 2026 | ||
| Visual Grounding | ScreenSpot | 20 | Apr 28, 2026 | ||
| Language Modeling | LAMBADA multilingual (test) | 20 | Apr 10, 2026 | ||
| Image Captioning | Flickr8k | 20 | Apr 10, 2026 | ||
| Model Editing | RuleEdit-200 | 20 | Apr 10, 2026 | ||
| Knowledge Tracing | XES3G5M | 20 | Apr 10, 2026 | ||
| Question Answering | NQ, TruthfulQA, WoW, HotpotQA, ELI5 | 20 | Apr 10, 2026 | ||
| Question Answering | HotpotQA | 20 | Apr 10, 2026 | ||
| Question Answering | WoW | 20 | Apr 10, 2026 | ||
| Question Answering | NQ | 20 | Apr 10, 2026 | ||
| Proactive Demand Detection | IntentFlow (Demand) | 20 | Apr 10, 2026 | ||
| Tabular Classification | URL | 20 | Apr 10, 2026 | ||
| Text-to-Image Generation | LAION-5B 1,000 prompts | 20 | Apr 10, 2026 | ||
| Multi-turn dialogue | TopDial | 20 | Apr 10, 2026 | ||
| Multi-turn dialogue | ConsistentChat | 20 | Apr 10, 2026 | ||
| Multi-turn dialogue | MT-Eval | 20 | Apr 10, 2026 | ||
| Classification | pop failures | 20 | Apr 10, 2026 | ||
| Remote Sensing Image Classification | LCZ HK | 20 | Apr 10, 2026 | ||
| Remote Sensing Image Classification | Yellow River Estuary | 20 | Apr 10, 2026 | ||
| Remote Sensing Image Classification | Augsburg | 20 | Apr 10, 2026 | ||
| Semantic Uncertainty Quantification | SQuAD (test) | 20 | Apr 10, 2026 | ||
| Semantic Uncertainty Quantification | NQ (test) | 20 | Apr 10, 2026 | ||
| Grade School Math Reasoning | GSM8K | 20 | Apr 10, 2026 |