Chaitanya Malaviya

I am a predoctoral young investigator at the Allen Institute for Artificial Intelligence, where I work on the Mosaic (Commonsense Reasoning) team with Professor Yejin Choi.

Before joining AI2, I completed my masters at the Language Technologies Institute, School of Computer Science at Carnegie Mellon University, where I was advised by Professor Graham Neubig as a member of Neulab. I did my bachelors at Nanyang Technological University, Singapore.

Email  /  CV  /  Google Scholar  /  Semantic Scholar  /  GitHub  /  Twitter

profile photo
Research Interests

I'm interested in robust machine learning methods for natural language processing. I also take a keen interest in linguistics for NLP and cognitive science. My research is centered around:

  1. devising training methods for improving model robustness,
  2. injecting world knowledge into reasoning models and
  3. introducing techniques to assess model robustness.

News

Dec 2019: 1 paper accepted to ICLR 2020.

Nov 2019: 1 paper accepted to AAAI 2020.

Aug 2019: New preprint on abductive commonsense reasoning.

May 2019: 1 paper accepted to ACL 2019.

Oct 2018: Joined the Allen Institute for AI as a predoctoral young investigator.

Publications
Commonsense Knowledge Base Completion with Structural and Semantic Context
Chaitanya Malaviya, Chandra Bhagavatula, Antoine Bosselut, Yejin Choi
AAAI, 2020  

BibTeX / Code

TL;DR: Fusing pre-trained language models and graph convolutional networks for commonsense knowledge graph completion.

Abductive Commonsense Reasoning
Chandra Bhagavatula, Ronan Le Bras, Chaitanya Malaviya, Keisuke Sakaguchi, Ari Holtzman, Hannah Rashkin, Doug Downey, Scott Wen-tau Yih, Yejin Choi
ICLR, 2020  

BibTeX

TL;DR: Abductive reasoning dataset and models for performing abductive reasoning in natural language.

COMET: Commonsense Transformers for Automatic Knowledge Graph Construction
Antoine Bosselut, Hannah Rashkin, Maarten Sap, Chaitanya Malaviya, Asli Celikylimaz, Yejin Choi
ACL, 2019  

BibTeX / Code / Demo

TL;DR: Transformer-based generative commonsense model for knowledge graph construction.

A Simple Joint Model for Improved Contextual Neural Lemmatization
Chaitanya Malaviya*, Shijie Wu*, Ryan Cotterell
NAACL, 2019  

BibTeX / Code / Slides / Talk

TL;DR: Jointly factorized model for morphological tagging and lemmatization.

Neural Factor Graph Models for Cross-lingual Morphological Tagging
Chaitanya Malaviya, Matthew R. Gormley, Graham Neubig
ACL, 2018  

BibTeX / Code / Slides / Talk

TL;DR: Neural Factor Graph: Factorial Conditional Random Field fused with LSTM potentials.

Sparse and Constrained Attention for Neural Machine Translation
Chaitanya Malaviya, Pedro Ferreira, André F.T. Martins
ACL, 2018  

BibTeX / Code / Slides / Talk

TL;DR: Sparse and Constrained Attention to improve adequacy of neural machine translation.

Learning Language Representations for Typology Prediction
Chaitanya Malaviya, Graham Neubig, Patrick Littell
EMNLP, 2017  

BibTeX / Code / Poster

TL;DR: Demystifying what neural MT learns about linguistic typology.

The SIGMORPHON 2019 Shared Task: Morphological Analysis in Context and Cross-Lingual Transfer for Inflection
Arya McCarthy, Ekaterina Vylomova, Shijie Wu, Chaitanya Malaviya, Lawrence Wolf-Sonkin, Garrett Nicolai, Miikka Silfverberg, Sabrina Mielke, Jeffrey Heinz, Ryan Cotterell, Mans Hulden
SIGMORPHON, 2019  

BibTeX

TL;DR: The SIGMORPHON 2019 Shared Task Report.

Technical Reports
Building CMU Magnus from User Feedback
Shrimai Prabhumoye, Fadi Botros, Khyathi Chandu, Samridhi Choudhary, Esha Keni, Chaitanya Malaviya, Thomas Manzini, Rama Pasumarthi, Shivani Poddar, Abhilasha Ravichander, Zhou Yu, Alan Black
Alexa Prize Proceedings, 2017  

BibTeX

TL;DR: System Description for Alexa Prize 2017 Chatbot.

DyNet: The Dynamic Neural Network Toolkit
Graham Neubig, Chris Dyer, Yoav Goldberg, Austin Matthews, Waleed Ammar, Miguel Ballesteros, David Chiang, Daniel Clothiaux, Trevor Cohn, Kevin Duh, Manaal Faruqui, Cynthia Gan, Dan Garrette, Yangfeng Ji, Lingpeng Kong, Adhiguna Kuncoro, Gaurav Kumar, Chaitanya Malaviya, Paul Michel, Yusuke Oda, Matthew Richardson, Naomi Saphra, Swabha Swayamdipta, Pengcheng Yin
arXiv, 2017  

BibTeX / Source

TL;DR: Techical Report for the DyNet neural network library.


Website source from Jon Barron here