ServiceNow Research

Dzmitry Bahdanau

Dzmitry Bahdanau

Research Lead

Human Machine Interaction Through Language

Dzmitry (Dima) Bahdanau is a Research Scientist and the Research Lead of the Human-Machine Interaction Through Language Program at ServiceNow Research. He holds a PhD in Computer Science from Mila, Université de Montréal, where he was supervised by Yoshua Bengio. Dzmitry’s research interests include but are not limited to goal-driven dialogue, AI-based code modeling and semantic parsing, language understanding and systematic generalization, connecting symbolic AI with deep learning, semantic parsing and goal-driven dialogue. In addition to his position at ServiceNow Research, Dzmitry is an Adjunct Professor at McGill University and Industry Core Member at Mila, Quebec Artificial Intelligence Institute.

Interests
  • Dialog
  • Large Language Models
  • Human Machine Interaction Through Language
  • Text2Code
  • Compositional Generalization
  • Natural Language Processing
  • Intent Classification
  • Symbolic Reasoning
  • Neurosymbolic AI
  • Semantic Parsing

Publications

SantaCoder: don't reach for the stars! . Harm de Vries, Raymond Li, Joel Lamy Poirier, Dzmitry Bahdanau, Denis Kocetkov, Sean Hughes. At ArXiv,  2022.

PDF Cite

On the Compositional Generalization Gap of In-Context Learning. Dzmitry Bahdanau, Arian Hosseini, Aaron Courville, Alessandro Sordoni, Ankit Vani. At Conference on Empirical Methods in Natural Language Processing (EMNLP),  2022.

Cite

Data Augmentation for Intent Classification with Off-the-shelf Large Language Models. Gaurav Sahu, Pau Rodriguez, Issam H. Laradji, Parmida Atighhehchian, David Vazquez, Dzmitry Bahdanau. At Montreal AI Symposium (MAIS),  2022.

PDF Cite

Data Augmentation for Intent Classification with Off-the-shelf Large Language Models. Gaurav Sahu, Pau Rodriguez, Parmida Atighhehchian, Issam H. Laradji, David Vazquez, Dzmitry Bahdanau. At Workshop at the Annual Meetings of the Association for Computational Linguistics (ACL),  2022.

PDF Cite

Compositional Generalization in Dependency Parsing. Emily Goodwin, Siva Reddy, Timothy J. O'Donnell, Dzmitry Bahdanau. At Annual Meeting of the Association for Computational Linguistics (ACL),  2022.

PDF Cite

LAGr: Labeling Aligned Graphs for Improving Systematic Generalization in Semantic Parsing. Dora Jambor, Dzmitry Bahdanau. At Annual Meeting of the Association for Computational Linguistics (ACL),  2022.

PDF Cite

Systematic Generalization with Edge Transformers. Leon Bergen, Timothy J. O'Donnell, Dzmitry Bahdanau. At Conference on Neural Information Processing Systems (NeurIPS),  2021.

PDF Cite

Picard: Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Models. Torsten Scholak, Nathan Schucher, Dzmitry Bahdanau. At Conference on Empirical Methods in Natural Language Processing (EMNLP),  2021.

PDF Cite Code Video

DuoRAT: Towards Simpler Text-to-SQL Models. Torsten Scholak, Raymond Li, Dzmitry Bahdanau, Harm de Vries, Christopher Pal. At North American Chapter of the Association for Computational Linguistics (NAACL),  2021.

PDF Cite Code

Understanding by Understanding Not: Modeling Negation in Language Models. Arian Hosseini, Siva Reddy, Dzmitry Bahdanau, R Devon Hjelm, Alessandro Sordoni, Aaron Courville. At North American Chapter of the Association for Computational Linguistics (NAACL),  2021.

PDF Cite

Towards Ecologically Valid Research on Language User Interfaces. Harm de Vries, Dzmitry Bahdanau, Chris Manning. At ArXiv,  2020.

PDF Cite

CLOSURE: Assessing Systematic Generalization of CLEVR models. Dzmitry Bahdanau, Harm de Vries, Timothy J. O'Donnell, Shikhar Murty, Philippe Beaudoin, Yoshua Bengio, Aaron Courville. At Workshop at the Neural Information Processing Systems (NeurIPS),  2019.

PDF Cite Code

BabyAI: A Platform to Study the Sample Efficiency of Grounded Language Learning. Maxime Chevalier-Boisvert, Dzmitry Bahdanau, Salem Lahlou, Lucas Willems, Chitwan Saharia, Thien Huu Nguyen, Yoshua Bengio. At International Conference on Learning Representations (ICLR),  2019.

PDF Cite Code

Systematic Generalization: What Is Required and Can It Be Learned?. Dzmitry Bahdanau, Shikhar Murty, Michael Noukhovitch, Thien Huu Nguyen, Harm de Vries, Aaron Courville. At International Conference on Learning Representations (ICLR),  2019.

PDF Cite