Researchers have introduced Chinese-SkillSpan, the first dataset specifically designed for extracting job-related skills from Chinese job advertisements. This dataset, aligned with the ESCO occupational skill standard, was created using a pipeline that combines LLM-powered annotation with expert review. The dataset contains over 20,000 annotated instances collected from major recruitment platforms between 2014 and 2025. Chinese-SkillSpan aims to address a gap in Chinese JobSkillNER resources and facilitate research in intelligent recruitment. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
IMPACT Provides a new benchmark for Chinese job skill extraction, potentially improving talent matching and recruitment platforms.
RANK_REASON Release of a new dataset for a specific NLP task (JobSkillNER) and its associated annotation methodology.