| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Language Modeling | PTB | Perplexity4.345 | 1,234 | |
| Language Modeling | PTB (test) | Perplexity8.159 | 543 | |
| Imputation | PTB | PRD4.16 | 162 | |
| Constituent parsing | PTB (test) | F196.48 | 155 | |
| Language Modeling | PTB (val) | Perplexity20.26 | 107 | |
| Unsupervised Parsing | PTB (test) | F1 Score83.3 | 75 | |
| Open-ended text generation | PTB | COH Score67.7 | 64 | |
| Phrase-structure parsing | PTB (§23) | F1 Score95.8 | 56 | |
| ECG Classification | PTB-XL (test) | AUC97.7 | 46 | |
| Character-level Prediction | PTB (test) | BPC (Test)1.045 | 42 | |
| Dependency Parsing | PTB | LAS96.3 | 31 | |
| Grammar Induction | PTB English (test) | F1 Score84.3 | 29 | |
| Time-series Classification | PTB 2 classes | Accuracy99.9 | 26 | |
| ECG Classification | PTB-XL | AUROC96 | 26 | |
| Language Modeling | PTB zero-shot | Perplexity82.05 | 25 | |
| POS tagging | PTB (test) | Accuracy97.78 | 24 | |
| Word-level prediction | PTB word-level (test) | Perplexity72.8 | 19 | |
| Parsing | PTB (test) | Sents/sec1,092 | 17 | |
| 2-class ECG classification | PTB ECG-2 (Cross-subject) | Accuracy87.73 | 13 | |
| Language Modeling | PTB English Mikolov preprocessed (val) | Perplexity44.9 | 13 | |
| Constituency Parsing | PTB Section 23 (test) | Rerank F10.9513 | 13 | |
| ECG Classification | PTB database | Accuracy99.43 | 13 | |
| Medical Time Series Classification | PTB 2-Classes | Accuracy85.39 | 12 | |
| Conditional Generation | PTB | Perplexity41.2 | 12 | |
| Uncertainty Quantification | PTB | CU417 | 12 |