Loading the SOTA2 catalog…
SOTA2 Research · benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
| Task | Dataset | Domain | Trend | Results | Last Update |
|---|---|---|---|---|---|
| Graph Fraud Detection | Yelp (test) | 40 | May 26, 2026 | ||
| Graph Fraud Detection | Amazon (test) | 40 | May 26, 2026 | ||
| Image Classification | Caltech101 | 40 | Jun 23, 2026 | ||
| Computational Complexity Analysis | WildGuardMix 1.0 (test) | 40 | May 26, 2026 | ||
| Safety Monitoring | WildGuardMix (test) | 40 | May 26, 2026 | ||
| Classification |
| Liver ultrasound dataset (10% train, 90% test) |
| 40 |
| May 26, 2026 |
| Semi-supervised classification | ADNI CN vs. AD (90% test) | 40 | May 26, 2026 |
|---|
| Long-context understanding | LongBench | 40 | Jun 11, 2026 |
|---|
| Generation | Coding domain | 40 | May 26, 2026 |
|---|
| Generation | QA domain | 40 | May 26, 2026 |
|---|
| Generation | Math domain | 40 | May 26, 2026 |
|---|
| Code Generation | Coding | 40 | May 26, 2026 |
|---|
| Question Answering | QA | 40 | May 26, 2026 |
|---|
| Safety and Utility Evaluation | Safety and Utility evaluation suite (test) | 40 | May 26, 2026 |
|---|
| Semi-supervised image classification | Corel5k 10% (train) | 40 | May 26, 2026 |
|---|
| Semi-supervised image classification | Flowers 10% (train) | 40 | May 26, 2026 |
|---|
| Zero-shot Reasoning and General Knowledge | Evaluation Suite Zero-shot (PIQA, ARC-Easy, ARC-Challenge, HellaSwag, WinoGrande, MathQA, MMLU) | 40 | May 25, 2026 |
|---|
| Text-to-Speech | Seed-TTS-Eval (test) | 40 | Jun 4, 2026 |
|---|
| Matrix Estimation | MovieLens Symmetric Noise | 40 | May 21, 2026 |
|---|
| Matrix Estimation | MovieLens Pairflip Noise | 40 | May 21, 2026 |
|---|
| Mathematical Reasoning | Mathematical Reasoning Suite GSM8K, MATH, SVAMP, SimulEq, AQuA, SAT, MMLU | 40 | May 21, 2026 |
|---|
| Question Answering | TriviaQA | 40 | May 20, 2026 |
|---|
| Selective Question Answering | TriviaQA | 40 | May 20, 2026 |
|---|
| Engineering Design Workflow Evaluation | Beams2D | 40 | May 20, 2026 |
|---|
| Faithfulness Evaluation | ImageNet-1K | 40 | May 19, 2026 |
|---|