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 |
|---|---|---|---|---|---|
| Source Code Watermark Extraction Robustness | CodeNet D40 (evaluation) | 36 | Jul 8, 2026 | ||
| Source Code Watermark Extraction Robustness | CodeNet D32 (evaluation) | 36 | Jul 8, 2026 | ||
| Watermark Detection Robustness | C4 | 36 | Jul 8, 2026 | ||
| Code Generation | HE+ | 36 | Jul 7, 2026 | ||
| Interactive Agent Task | ALFWorld |
| 36 |
| Jul 7, 2026 |
| Code Generation | LiveCodeBench | 36 | Jul 3, 2026 |
|---|
| Six-class anatomical-region classification | Mayo 2020 (test) | 36 | Jul 3, 2026 |
|---|
| Question Answering | 2WikiMultihopQA | 36 | Jul 2, 2026 |
|---|
| Question Answering | MuSiQue | 36 | Jul 2, 2026 |
|---|
| Question Answering | HotpotQA | 36 | Jul 2, 2026 |
|---|
| Multihop Question Answering | 2WikiMultihopQA | 36 | Jul 2, 2026 |
|---|
| Multihop Question Answering | MuSiQue | 36 | Jul 2, 2026 |
|---|
| Multihop Question Answering | HotpotQA | 36 | Jul 2, 2026 |
|---|
| LLM Prefill throughput | LLM Prompt Length 2048 (pp2048) | 36 | Jul 2, 2026 |
|---|
| LLM Prefill throughput | LLM Prompt Length 1024 | 36 | Jul 2, 2026 |
|---|
| Multiple choice question visual question answering | Multiple choice question (MCQ) visual question answering (VQA) benchmarks Average | 36 | Jul 1, 2026 |
|---|
| Image Classification | Aircraft | 36 | Jul 1, 2026 |
|---|
| Image Classification | StanfordCars | 36 | Jul 1, 2026 |
|---|
| Image Classification | Pets | 36 | Jul 1, 2026 |
|---|
| Mathematical Reasoning | GSM8K | 36 | Jul 7, 2026 |
|---|
| Downstream AUROC performance evaluation | Camelyon 17 | 36 | Jun 30, 2026 |
|---|
| Covariance Test-Time Adaptation | Terra | 36 | Jun 30, 2026 |
|---|
| Covariance Test-Time Adaptation | VisDA 2017 | 36 | Jun 30, 2026 |
|---|
| Language Model Training Efficiency | FineWeb | 36 | Jun 30, 2026 |
|---|
| Localization | Blackbird clov, patr, thri, wint, amp, dice, half, sid trajectories (test) | 36 | Jun 30, 2026 |
|---|