Reproduce “CSWBench: Evaluating Context-Sensitive Code-Switching in Creative Writing” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Zewen Jiang, Yahan Li, Chenlu Guo, Diana Inkpen, Yuan Wu, Yi Chang
CSWBench evaluates code-switching in Chinese-dominant narratives with Latin-script English, assessing switch events and multilingual organization. It combines local judgments of whether switches are licensed, appropriate, and functional with whole-text ratings, and tests placement by swapping switch events or Chinese spans in 36 Extended narratives. Both swaps lowered ratings, with different dimension-specific effects; real-world bilingual interaction remains outside its scope.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.