Empowering a Metric with LLM-assisted Named Entity Annotation: HW-TSC's Submission to the WMT23 Metrics Shared Task.
Zhanglin WuYilun LiuMin ZhangXiaofeng ZhaoJunhao ZhuMing ZhuXiaosong QiaoJingfei ZhangMiaomiao MaYanqing ZhaoSong PengShimin TaoHao YangYanfei JiangPublished in: WMT (2023)
Keyphrases
- named entities
- named entity recognition
- annotated corpus
- evaluation metrics
- information extraction
- relation extraction
- natural language processing
- co occurrence
- named entity extraction
- text mining
- coreference resolution
- question answering
- noun phrases
- semantic role labeling
- automatic annotation
- semantic annotation
- metadata
- semi supervised
- linguistic features
- maximum entropy
- conditional random fields
- maximum entropy model
- unsupervised learning
- active learning
- data mining
- pairwise
- feature extraction
- knowledge base
- artificial intelligence