Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1805 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Homography Estimation | CAHomo (test) | 18 | Jun 5, 2026 | ||
| Dense Camera Motion Estimation | GHOF-Cam (test) | 18 | Jun 5, 2026 | ||
| Web-based Task Automation | WebShop (test) | 18 | Jun 5, 2026 | ||
| Interactive Task Learning | ALFWorld (val_seen) | 18 | Jun 5, 2026 | ||
| Mathematical Reasoning | MATH | 18 | Jun 5, 2026 | ||
| Mathematical Reasoning | GSM8K | 18 | Jun 5, 2026 | ||
| Cross-Subject Emotion Prediction Regression | SEED-IV Cross-Subject | 18 | Jun 5, 2026 | ||
| Emotion Prediction Regression | Long-Term Naturalistic Emotion (LTNE) Dataset Dominance | 18 | Jun 5, 2026 | ||
| Continuous Emotion Regression | Long-Term Naturalistic Emotion Dataset Arousal (Cross-Subject) | 18 | Jun 5, 2026 | ||
| Emotion Prediction Regression | SEED (Cross-Subject) | 18 | Jun 5, 2026 | ||
| Skill-selection | ToolBench-60 soft-0.3 preference regime (held-out pool) | 18 | Jun 5, 2026 | ||
| Agent Security Evaluation | Agent Security Bench (ASB) IPI Combined Attack (CA) and Context Ignoring (CI) averaged | 18 | Jun 5, 2026 | ||
| Agent Security Evaluation | Agent Security Bench (ASB) DPI Combined Attack (CA) and Context Ignoring (CI) averaged | 18 | Jun 5, 2026 | ||
| Scientific Question Answering | GPQA | 18 | Jun 5, 2026 | ||
| Mathematical Reasoning | MATH 500 | 18 | Jun 5, 2026 | ||
| Mathematical Reasoning | GSM8K | 18 | Jun 5, 2026 | ||
| Mathematical Reasoning | AIME 25 | 18 | Jun 5, 2026 | ||
| Mathematical Reasoning | AIME 24 | 18 | Jun 5, 2026 | ||
| General Language Understanding and Reasoning | General Benchmarks MMLU, CEval, CMMLU, HumanEval, MBPP, MATH, GSM8K, MultiArith, BBH | 18 | Jun 5, 2026 | ||
| Logical Reasoning | BBH | 18 | Jun 30, 2026 | ||
| Poetry Generation | Poetry | 18 | Jun 5, 2026 | ||
| Continual Visual Instruction Tuning | CVIT | 18 | Jun 5, 2026 | ||
| Language Modeling | WikiText-103 Clean (val) | 18 | Jun 5, 2026 | ||
| Natural Language Understanding | Ten Benchmark Datasets MRPC, QQP, STS-B, MNLI, QNLI, RTE, SNLI, Sci, SICK, Twi | 18 | Jun 5, 2026 | ||
| Collaborative filtering | MovieLens 100k | 18 | Jun 5, 2026 |