I'm a first-year PhD student at UMass-Amherst with a focus in deep learning and natural language processing.



Mar 2019
Paper submission to NAACL-2019 has been accepted and chosen for oral presentation.
Feb 2019
Excited to be interning at Google AI Language this summer.
Sep 2018
I've started my PhD at UMass-Amherst. Fortunate (and thrilled) to join IESL and be advised by Prof. Andrew McCallum.


Main Conference
Andrew Drozdov, Pat Verga, Mohit Yadav, Mohit Iyyer, Andrew McCallum
Unsupervised Latent Tree Induction with Deep Inside-Outside Recursive Auto-Encoders
Accepted as Oral at NAACL 2019
Pre-print not yet available.
Katrina Evtimova, Andrew Drozdov, Douwe Kiela, Kyunghyun Cho
Emergent Communication in a Multi-Modal, Multi-Step Referential Game
Accepted as Poster at ICLR 2018
pdf, source
Adina Williams, Andrew Drozdov, Samuel Bowman
Learning to parse from a semantic objective: It works. Is it syntax?
Accepted for Publication in TACL 2018
pdf, source
Andrew Drozdov and Samuel Bowman
The Coadaptation Problem when Learning How and What to Compose
Accepted as Extended Abstract at ACL 2017
Unpublished Work
Katrina Evtimova and Andrew Drozdov (Advised by David Sontag)
Understanding Mutual Information and its use in InfoGAN
Course Paper for Inference and Representation (2016)
pdf, award
Sébastien Jean, Andrew Drozdov, and Arpit Jain
POS Tagging and Chunking with Subword2Word Models
Course Paper for Statistical Natural Language Processing (2015)