Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1626 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Robustness against harmful content generation | LMSYS-MF attacks | 20 | Feb 26, 2026 | ||
| Memorization Reduction | GSM8K | 20 | Feb 26, 2026 | ||
| Memorization Reduction | Ultrachat | 20 | Feb 26, 2026 | ||
| N-1 Reliability Analysis | N-1 (test) | 20 | Feb 26, 2026 | ||
| Physical Reasoning | PIQA | 20 | Feb 26, 2026 | ||
| Cloze-style completion | Lambada OpenAI | 20 | Feb 26, 2026 | ||
| Seizure Onset Zone Localization | Epilepsy-iEEG-Multicenter (Patient-wise) | 20 | Feb 26, 2026 | ||
| Synthetic Sequence Modeling | Synthetic Tasks Dataset (test) | 20 | Feb 26, 2026 | ||
| Irregular Time Series Classification | MIMIC-III | 20 | Apr 15, 2026 | ||
| Irregular Time Series Classification | PhysioNet | 20 | Apr 15, 2026 | ||
| RAG Question Answering | Musique | 20 | Feb 26, 2026 | ||
| RAG Question Answering | NQ (Natural Questions) | 20 | Feb 26, 2026 | ||
| RAG Question Answering | HotpotQA | 20 | Feb 26, 2026 | ||
| Sequential Recommendation | Instrument | 20 | Feb 26, 2026 | ||
| Short Video Recommendation | RecIF-Bench Short Video Rec | 20 | Feb 26, 2026 | ||
| Recommendation | Goodreads | 20 | May 11, 2026 | ||
| Sequential Recommendation | Instruments | 20 | Mar 30, 2026 | ||
| Multi-hop Question Answering | HotpotQA | 20 | Feb 26, 2026 | ||
| Question Answering | HotpotQA | 20 | Feb 26, 2026 | ||
| Reciprocal Recommendation | Speed Dating Experiment Data | 20 | Feb 26, 2026 | ||
| Sequential Recommendation | Toys | 20 | Feb 26, 2026 | ||
| Retrieval | Overall (Average) | 20 | Jun 23, 2026 | ||
| Top-K Recommendation | Beauty (test) | 20 | Jun 1, 2026 | ||
| Information Retrieval | NQ BEIR | 20 | Feb 26, 2026 | ||
| Multimodal Recommendation | Arts, Crafts & Sewing | 20 | Feb 26, 2026 |