ServiceNow Research

Harm de Vries

Harm de Vries

Research Lead

Large Language Models Lab

Harm de Vries is a Research Scientist in the Human-Machine Interaction Through Language program at ServiceNow Research. He holds a PhD in Computer Science from Mila, Universite de Montreal, where he was supervised by Aaron Courville. During his PhD, he worked on machine learning methods for (visually) grounded language understanding. He’s currently focused on how NLP methods can improve human-machine interaction and is therefore interested in a wide range of research topics such as large language models, semantic parsing, task-oriented dialogue, and text2code. His research has been published in EMNLP, ICLR, NeurIPS, CVPR, and ECCV.

  • Text2Code
  • Semantic Parsing
  • Large Language Models


SantaCoder: don't reach for the stars!. Workshop at the International Conference on Learning Representations (ICLR),  2023.

PDF Cite

The Power of Prompt Tuning for Low-Resource Semantic Parsing. Annual Meeting of the Association for Computational Linguistics (ACL),  2022.

PDF Cite

DuoRAT: Towards Simpler Text-to-SQL Models. North American Chapter of the Association for Computational Linguistics (NAACL),  2021.

PDF Cite Code