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 |
|---|---|---|---|---|---|
| Watermark Robustness | VCTK, LibriSpeech, and LJSpeech (test) | 36 | Jun 11, 2026 | ||
| westernized (2 classes) classification | cMD3 | 36 | Jun 11, 2026 | ||
| Language Understanding | GLUE and SuperGLUE fine-tuning | 36 | Jun 11, 2026 | ||
| Vehicle Routing Problem | 8 Unseen VRP Variants 1.0 (test) | 36 | Jun 11, 2026 | ||
| Model diagnostic efficiency | LOMO behavioral dataset (held-out model fold) |
| 36 |
| Jun 11, 2026 |
| Knowledge-intensive Question Answering | CRAG | 36 | Jun 11, 2026 |
|---|
| Bag-level classification | NCT-CRC | 36 | Jun 11, 2026 |
|---|
| Instance-level classification | NCT-CRC | 36 | Jun 11, 2026 |
|---|
| Bimanual Manipulation | RoboTwin Clean setting 2.0 | 36 | Jun 16, 2026 |
|---|
| Imputation | OpinionQA MNAR 140 opinion variables (W26 to W82) American Trends Panel | 36 | Jun 9, 2026 |
|---|
| Imputation | OpinionQA MCAR 140 opinion variables (W26 to W82) American Trends Panel | 36 | Jun 9, 2026 |
|---|
| Hypergraph Clustering | Contact-high-school | 36 | Jun 9, 2026 |
|---|
| Canary Leakage Evaluation | Code Canary (train) | 36 | Jun 9, 2026 |
|---|
| Canary Leakage Evaluation | Prompt Canary (train) | 36 | Jun 9, 2026 |
|---|
| Canary Leakage Evaluation | Joint Canary (train) | 36 | Jun 9, 2026 |
|---|
| Jailbreaking Tool-Using LLM Agents | AgentDojo | 36 | Jun 9, 2026 |
|---|
| Symbolic Planning | PDDL | 36 | Jun 9, 2026 |
|---|
| Node Clustering | PubMed | 36 | Jun 9, 2026 |
|---|
| Node Clustering | CiteSeer | 36 | Jun 9, 2026 |
|---|
| Node Clustering | Cora | 36 | Jun 9, 2026 |
|---|
| Pragmatic Reasoning | PRAGMEGA (test) | 36 | Jun 9, 2026 |
|---|
| Backdoor Detection | Backdoor behaviors (Jailbreak, Watching, Refusal, Password, Country, Sentiment) zero-shot | 36 | Jun 9, 2026 |
|---|
| Passenger queue length forecasting | Incheon International Airport (ICN) Departure Gates (test) | 36 | Jun 9, 2026 |
|---|
| Emotion Recognition | DEAP | 36 | Jul 7, 2026 |
|---|
| Three-objective alignment | Helpful Assistant (test) | 36 | Jun 8, 2026 |
|---|