Hi! I’m a PhD student at the Language Technologies Institute in the School of Computer Science at Carnegie Mellon working with Daniel Fried. I like to think about how we can communicate with machines using natural language, code, demonstrations, and other means to specify the tasks we want them to do. I’m also broadly interested in natural language processing, machine learning, and program synthesis, especially in how these fields interact with linguistics and cognitive science.

I completed my undergraduate and master’s degrees at IIIT Hyderabad studying computer science and computational linguistics, advised by Monojit Choudhury and Dipti Misra Sharma. I have spent at Chandar Research Lab at Mila (with Sarath Chandar and Prasanna Parthasarathi).

Feel free to get in touch if you want to chat about research! If you’re looking to apply to PhD programs in NLP or ML, I annotated my statement of purpose with the ideas and reasoning that went behind it in the hope that it will demystify the process for others. Check it out along with other resources at this page!

News

20 May 2024Starting an internship at Autodesk working with Evan Pu for the summer!
15 May 2024Akhila's paper on evaluating non-literal language understanding in conversational agents has been accepted to appear at ACL 2024!
1 May 2024Our paper on amortizing pragmatic program synthesis with ranking has been accepted to appear at ICML!
16 January 2024Our paper on training pragmatic program synthesizers has been accepted to appear at ICLR 2024!
9 November 2023The paper I led on using pragmatics to train program synthesizers is up on arXiv! Super excited to be able to share this work.

Publications

Amortizing Pragmatic Program Synthesis with Rankings
Yewen Pu, Saujas Vaduguru, Priyan Vaithilingam, Elena Glassman, Daniel Fried
41st International Conference on Machine Learning (ICML), 2024.

Is the Pope Catholic? Yes, the Pope is Catholic. Generative Evaluation of Intent Resolution in LLMs
Akhila Yerukola, Saujas Vaduguru, Daniel Fried, Maarten Sap
62nd Annual Meeting of the Association for Computational Linguistics (ACL), 2024.

Generating Pragmatic Examples to Train Neural Program Synthesizers
Saujas Vaduguru, Daniel Fried, Yewen Pu
12th International Conference on Learning Representations (ICLR), 2024.

Symbolic Planning and Code Generation for Grounded Dialogue
Justin T. Chiu, Wenting Zhao, Derek Chen, Saujas Vaduguru, Alexander M. Rush, Daniel Fried
Conference on Empirical Methods in Natural Language Processing (EMNLP), 2023.

Adapting to Gradual Distribution Shifts with Continual Weight Averaging
Jared Fernandez, Saujas Vaduguru, Sanket Vaibhav Mehta, Yonatan Bisk, Emma Strubell
Workshop on High-dimensional Learning Dynamics
40th International Conference on Machine Learning (ICML), 2023.

Probing Negation in Language Models
Shashwat Singh, Shashwat Goel, Saujas Vaduguru, Ponnurangam Kumaraguru
8th Workshop on Representation Learning for NLP (RepL4NLP)
61st Annual Meeting of the Association for Computational Linguistics (ACL), 2023.

Stress Rules from Surface Forms
Saujas Vaduguru, Partho Sarthi, Dipti Misra Sharma, Monojit Choudhury
International Conference on Natural Language Processing (ICON), 2021.

Efficient Pragmatic Program Synthesis with Informative Specifications
Saujas Vaduguru, Kevin Ellis, Yewen Pu
Workshop on Meaning in Context: Pragmatic Communication in Humans and Machines
35th Conference on Neural Information Processing Systems (NeurIPS), 2021.

[talk], [code]

Sample-efficient Linguistic Generalizations through Program Synthesis: Experiments with Phonology Problems
Saujas Vaduguru, Aalok Sathe, Monojit Choudhury, Dipti Misra Sharma
Proceedings of the 18th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology
The Joint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (ACL-IJCNLP), 2021.

[dataset]