Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1260 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Dialogue Summarization | DialogSum | 24 | May 22, 2026 | ||
| Object Detection | COCO-UniFS (val) | 24 | Feb 27, 2026 | ||
| Action Recognition | NTU RGB+D 120 (110/10) | 24 | Apr 2, 2026 | ||
| Toxicity classification | ToxCMM | 24 | Feb 26, 2026 | ||
| Healthcare Record Management | EHRAgent | 24 | Feb 26, 2026 | ||
| Knowledge-intensive QA | StrategyQA | 24 | Feb 26, 2026 | ||
| Autonomous Driving | Agent-Driver | 24 | Feb 26, 2026 | ||
| Bi-directional MR model estimation | Synthetic bi-directional MR data Scenario S(4,4,10) direction Y → X | 24 | Feb 26, 2026 | ||
| Bi-directional MR model estimation | Synthetic bi-directional MR data Scenario S(2,2,6) direction X → Y | 24 | Feb 26, 2026 | ||
| Medical Image Classification | Chest Xray | 24 | May 7, 2026 | ||
| Web Task | WebShop | 24 | Feb 26, 2026 | ||
| Fact Verification | LIAR | 24 | Apr 8, 2026 | ||
| Node Classification | Texas Hnode = 0.11 | 24 | Feb 26, 2026 | ||
| Instruction Following Evaluation | Average (Vicuna, Self-instruct, Dolly, BPO) (test) | 24 | Feb 26, 2026 | ||
| Discriminative Evaluation | CWQ (test) | 24 | Feb 26, 2026 | ||
| Dialogue | 4 dialogue tasks (Skill Talk, Empathetic Dialogues, Wizard of Internet, Wizard of Wikipedia) (test) | 24 | Feb 26, 2026 | ||
| Classification | 9 classification tasks (test) | 24 | Feb 26, 2026 | ||
| Machine Unlearning | 32 target sequences (unlearning set) | 24 | Feb 26, 2026 | ||
| Blackjack | Blackjack 900 games | 24 | Feb 26, 2026 | ||
| Personality Measurement | The 16 Personalities (test) | 24 | Feb 26, 2026 | ||
| Personality Measurement | The Big Five Inventory (test) | 24 | Feb 26, 2026 | ||
| Hallucination Evaluation | MME hallucination (test) | 24 | Feb 26, 2026 | ||
| Reinforcement Learning | MuJoCo HumanoidStandup | 24 | Feb 26, 2026 | ||
| Object Detection | SIM10k -> Cityscapes | 24 | Jul 1, 2026 | ||
| LLM Generation | RTX 3090 24GB (inference) | 24 | Feb 26, 2026 |