| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| 60-task evaluation suite 10 domains | Code PQS89 | 19 | 2mo ago | ||
| Robo Challenge (Online) | State-Aware CP-SAT Repair | Plan Accuracy85.7 | 16 | 1d ago | |
| Robo Challenge (Offline) | CP-SAT Formalizer | Plan Accuracy100 | 12 | 1d ago | |
| RecipeNLG (test) | Self-Ask | BLEU91.3 | 7 | 3mo ago | |
| WikiHow (test) | Self-Ask | ROUGE-156.1 | 7 | 3mo ago | |
| 6 grid world domains Unseen Appearances | VLMFP | Success Rate (Frozenlake)81.1 | 6 | 2mo ago | |
| 6 grid world domains Seen Appearances | VLMFP | Success Rate (Frozenlake)95.2 | 6 | 2mo ago | |
| Commonly solved tasks | DeepSeek | Runtime Ratio0.36 | 4 | 2mo ago |
