CLUE's Traditional-Chinese span-extraction task, adopted unchanged from the separately published Delta Reading Comprehension Dataset; not scored on CLUE's live leaderboard.
unassessed
| Category | composite |
|---|---|
| Subcategory | span-extraction reading comprehension (Traditional Chinese) |
| Page status | active |
| Metric | exact match (EM); F1 also reported by the original paper |
| Direction | higher_is_better |
| Unit | % |
| Dataset size | 33953 |
| Publisher | Delta Research Center -- the dataset's Chinese name (台達閱讀理解資料集) identifies it as Delta Electronics'; no further institutional detail was confirmed from the sources read |
DRCD gives a model a Traditional Chinese Wikipedia paragraph and a question, requiring an extracted answer span -- the same SQuAD-style span-extraction format as CMRC2018, but in Traditional rather than Simplified characters, and drawn from an entirely separate dataset. CLUE adopted DRCD unchanged from the Delta Reading Comprehension Dataset, published by a Delta Research Center team with no author overlap with the CLUE team -- the cleanest case of external adoption in this batch. CLUE provides its own traditional-to-simplified conversion tool for models evaluated in simplified Chinese but does not otherwise alter the data.
Extractive question answering: given a Traditional Chinese passage and question, output the answer text span; the reference OpenCompass config scores it by exact match (EM), and the original paper additionally reports F1.
No model card in ModelSpec reports this benchmark yet.