Chinese
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Joint Word Segmentation and POS Tagging | Chinese (test) | F1 Score84.5 | 36 | |
| Disease Classification | Chinese (external val) | AUROC100 | 18 | |
| Unsupervised Constituency Parsing | Chinese (test) | SF153.92 | 7 | |
| Multilingual Language Understanding | Chinese | Average Performance68.7 | 5 | |
| Speaker Diarization | Chinese Hard | DER10.18 | 5 | |
| Speaker Diarization | Chinese | DER8.325 | 5 | |
| Reference-based Quality Estimation | Chinese (ZH) | R_pb Score0.84 | 5 | |
| Zero-shot Text-to-Speech | Chinese Speech Emotion Prompt | WER0.0162 | 4 | |
| Named Entity Recognition | Chinese (test) | F1 Score68.59 | 4 | |
| Tokenization | Chinese | Average Tokens per Sample914.05 | 3 | |
| Vector Font Reconstruction | Chinese CN (test) | Error8 | 3 | |
| Definition Generation | Chinese (test) | Accuracy (a1)3 | 3 | |
| Simple Definition Generation | Chinese (test) | L1-3 Rate48.03 | 2 |
Showing 13 of 13 rows