Hi, I am a research scientist at Google AI Language. My research is focused on developing methods for neural networks that con generate natural language. Specifically, I want to put users of these models first by ensuring that they are controllable, robust, fair, and interpretable.
I like to work on large multi-disciplinary collaborations, for example the GEM benchmark for language generation, evaluation, and its metrics for which we are organizing a workshop at ACL 2021. Please consider participating!
Besides this, I also gave a tutorial on interpretability in NLP at ACL 2020, and co-organized a workshop on evaluating NLG models at INLG 2020.
Before joining Google, I completed my Ph.D. in Computer Science at Harvard, supervised by Barbara Grosz and Sasha Rush. My dissertation was titled “Human-AI Collaboration for Natual Language Generation with Interpretable Neural Networks”.
Publications
2021
Accelerating Antimicrobial Discovery with Controllable Deep Generative Models and Molecular Dynamics
Nature Biomedical Engineering
[Paper]
2020
Proceedings of the 1st Workshop on Evaluating NLG Evaluation
Workshop at INLG 2020
[Workshop Website]
[Proceedings]
Learning to Evaluate Translation Beyond English: BLEURT Submissions to the WMT Metrics 2020 Shared Task
WMT 2020
[Paper]
The Language Interpretability Tool: Extensible, Interactive Visualizations and Analysis for NLP Models
EMNLP Demo 2020
[Paper]
[Website]
[Tutorial]
Investigating gender bias in language models using causal mediation analysis
NeurIPS 2020
[Extended Preprint]
[Paper]
Parent perspectives in shared decision-making for children with medical complexity
Academic Pediatrics
[Paper (Paywall)]
A Corpus for Detecting High-Context Clinical Indications in Intensive Care Patient Notes Focusing on Frequently Readmitted Patients
Proceedings of The 12th Language Resources and Evaluation Conference.
[Paper]
exBERT: A Visual Analysis Tool to Explore Learned Representations in Transformer Models
ACL Demos 2020
[Paper]
2019
Visual Interaction with Deep Learning Models through Collaborative Semantic Inference
IEEE Transactions on Visualization and Computer Graphics (VAST 2019)
[Paper]
GLTR: Statistical Detection and Visualization of Generated Text
Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics: System Demonstrations
Nominated for Best Demo
[Paper]
[Demo]
[Code]
LSTM Networks Can Perform Dynamic Counting
Deep Learning and Formal Languages Workshop at ACL 19
[Paper]
Evaluating an Automated Mediator for Joint Narratives in a Conflict Situation
Behaviour & Information Technology
[Paper]
Improving Human Text Comprehension through Semi-Markov CRF-based Neural Section Title Generation
Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics
[Paper]
Interactive Visual Exploration of Latent Space (IVELS) for Peptide Auto-Encoder Model Selection
ICLR 2019 Workshop on Deep Generative Models for Highly Structured Data
[Paper]
Generating Abstractive Summaries with Finetuned Language Models
Proceedings of the 12th International Conference on Natural Language Generation (Challenge Track)
[Paper]
Margin Call: an Accessible Web-based Text Viewer with Generated Paragraph Summaries in the Margin
Proceedings of the 12th International Conference on Natural Language Generation (Demo Track)
[Paper]
2018
Debugging Sequence-to-Sequence Models with Seq2Seq-Vis
BlackboxNLP workshop at EMNLP 2018
[Paper]
[Demo]
End-to-End Content and Plan Selection for Natural Language Generation
Proceedings of the 11th International Conference on Natural Language Generation
Best ROUGE/METEOR/CIDEr
[Paper]
[Code]
[Challenge Website]
Towards Controllable Generation of Diverse Natural Language
Proceedings of the 11th International Conference on Natural Language Generation (E2E NLG challenge track)
[Paper]
[Challenge Website]
Seq2Seq-Vis: A Visual Debugging Tool for Sequence-to-Sequence Models
IEEE Transactions on Visualization and Computer Graphics (VAST 2018)
Honorable Mention
[Paper]
[Demo]
[Code]
Comparing deep learning and concept extraction based methods for patient phenotyping from clinical narratives
PLoS One
[Paper]
[Code]
[Data]
2017
LSTMVis: A Tool for Visual Analysis of Hidden State Dynamics in Recurrent Neural Networks
IEEE Transactions on Visualization and Computer Graphics (InfoVis 2017)
[Paper]
[Demo]
[Code]
Behind the Scenes: A Medical Natural Language Processing Project
International Journal of Medical Informatics
[Paper]
2015
Deploying AI Methods to Support Collaborative Writing: a Preliminary Investigation
Proceedings of the 33rd Annual ACM Conference, Extended Abstracts on Human Factors in Computing Systems
[Paper]