Name-based gender inference is a common tool in scientometrics, but it often fails for non-Western scholars. A new arXiv preprint, "Bridging the Language Gap in Scholarly Data II: Enhancing Gender Inference for Chinese Names," takes on this problem specifically for Chinese names.
The authors note that Chinese names appear in two forms—Hanzi and romanized—which likely complicates inference. The paper is positioned as the second in a series and is framed as a step toward more inclusive research evaluation.
The abstract is brief and does not yet describe the proposed method or results in detail. Still, the problem is clear: without better tools, bibliometric analyses may misgender or exclude Chinese researchers.