This model was obtained by instruction-tuning the corresponding ClueAI/PromptCLUE-base-v1-5 model on the spelling error corpus. — Language(s) (NLP): Chinese — Pretrained from model: ClueAI/PromptCLUE-base-v1-5 — Pretrained by dataset: 1M UDN news corpus — Finetuned by dataset: shibing624/CSC spelling error corpus (CN + TC) — Repository: https://github.com/TedYeh/ChinesespellingCorrection — Chinese spelling error correction task(SIGHAN2015): — FPR: False Positive Rate CodeTed/CGEDit — Chinese Grammatical Error Diagnosis by Task-Specific Instruction Tuning
Модальности:
Генерация текста
Задача: Генерация текста
Автор: CodeTed
Теги: t5, text2text-generation, CSC, CGED, spelling error, zh, text-generation-inference, endpoints_compatible
Лайков: 2 | Загрузок: 87
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.