Dr. Sarah Chen

Associate Professor of Computer Science

Stanford University

My research focuses on natural language processing, machine learning, and the intersection of language understanding with knowledge representation.

Cross-Lingual Semantic Parsing with Minimal Supervision

Oral Presentation — EMNLP 2025, Punta Cana, Dominican Republic — November 2025

This oral presentation covers our work on XSP-Transfer, a cross-lingual transfer method for semantic parsing that achieves strong results with as few as 50 annotated examples in the target language. The talk presents the core technical contribution—a language-agnostic alignment objective combined with structure-aware code-switching augmentation—and demonstrates its effectiveness across 10 typologically diverse languages on the Mschema2QA and MTOP benchmarks.

A key focus of the presentation is the practical implications for making NLP technology accessible to speakers of low-resource languages. We discuss how our few-shot approach could enable rapid deployment of semantic parsing systems for new languages, and present a case study of deploying XSP-Transfer for Yoruba and Swahili question-answering applications. The talk concludes with a discussion of remaining challenges, including handling morphologically complex languages and adapting to cultural differences in how questions are formulated.