I’m a PhD student in Computer Science at UMass Amherst, advised by Prof. Andrew McCallum. Previously, I completed my MS in Computer Science at Stanford, advised by Prof. Chris Manning, and completed my Bachelor’s thesis at the Institute for Natural Language Processing in Stuttgart, supervised by PD Dr. Sabine Schulte im Walde.
Most recently, I worked on deep learning for dialogue at Google Research. Before that, I worked on learning word vectors that encoded challenging lexical relations, especially hypernymy, and then interrogated what they actually did encode with VecEval.
Currently, I’m working on problems in discourse structure at the document level. I’m particularly interested in understanding the interactions that take place during peer review through the lens of discourse. My hope is that this work will contribute to making peer review more transparent and equitable.
Here is my resume.
To Plan or not to Plan? Discourse planning in slot-value informed sequence to sequence models for language generation
Neha Nayak, Dilek Hakkani-Tur, Marilyn Walker, Larry Heck
INTERSPEECH 2017 (2017)
[paper] [bibtex] [poster]
Evaluating Word Embeddings Using a Representative Suite of Practical Tasks
Neha Nayak, Gabor Angeli, Chris Manning
First Workshop on Evaluating Vector Space Representations for NLP (RepEval). 2016
[paper] [bibtex] [code] [slides]
Reports and unpublished work
Building a Conversational Agent Overnight with Dialogue Self-Play
Pararth Shah, Dilek Hakkani-Tür, Gokhan Tür, Abhinav Rastogi, Ankur Bapna, Neha Nayak, Larry Heck
arxiv preprint, 2018