HillZhang1999 / MuCGEC

MuCGEC中文纠错数据集及文本纠错SOTA模型开源;Code & Data for our NAACL 2022 Paper "MuCGEC: a Multi-Reference Multi-Source Evaluation Dataset for Chinese Grammatical Error Correction"
https://aclanthology.org/2022.naacl-main.227/
Apache License 2.0
509 stars 64 forks source link

demo 接口 #17

Closed lichuanxiang closed 2 years ago

lichuanxiang commented 2 years ago

想问下,http://139.224.234.18:5002/ ,这个 demo 中的【语法纠错(母语)】接口,是以下哪种策略? 1)直接调用 seq2edit 模型;2)直接调用 seq2seq 模型;3)seq2edit + seq2seq 做 merge;4)seq2edit + seq2seq + 规则

HillZhang1999 commented 2 years ago

直接调用母语的Seq2Edit模型