Omid Bakhshandeh

Research Scientist at Institute for Human and Machine Cognition

I recently defended my PhD under James Allen in Conversational Interaction and Spoken Dialogue Research Group at University of Rochester.


Room 627, Computing Studies Building,
University of Rochester,
Rochester, New York

Email: omidb {at}

Short Biography

Omid Bakhshandeh is a Research Scientist at Institute for Human and Machine Cognition (IHMC). He obtained his Ph.D. under James F. Allen in Conversational Interaction and Dialogue research group at the university of Rochester. His main research interest is the intersection of natural language understanding and machine learning with a special focus on learning meaning representation and common sense reasoning. He spent summer 2015 and summer 2016 at the IHMC and Adobe Research, respectively, working on machine reading and vision and language. Before the graduate school, as an undergrad researcher, his primary work was on multi-agent systems and small-size mobile robots where he was leading AI efforts at Mechatronics Research Lab.

I'm interested in building intelligent agents capable of performing human-level intelligent tasks. I am mainly focused on deep natural language understanding, with the ultimate goal of building systems which have inferential abilities for interacting with the world via natural language. These days I work on developing computational models of semantics for understanding different aspects of language.

Publication Highlights

Effective Broad-Coverage Deep Parsing
James Allen, Omid Bakhshandeh, Choh Man Teng, Lucian Galescu, Will de Beaumont
AAAI 2018
Proposing Plausible Answers for Open-ended Visual Question Answering
Omid Bakhshandeh , Trung Bui, Zhe Lin, Walter Chang
arXiv preprint 2016
Learning to Jointly Predict Ellipsis and Comparison Structures
Omid Bakhshandeh , Alexis Welwood, James Allen
CoNLL 2016 [bib]
Formal Verification of Temporal Questions In the Context of Query-Answering Text Summarization
Nasrin Mostafazadeh, Omid Bakhshandeh , Gholamreza Ghassem-Sani
CAAI 2012 [bib]
Bayer-Based Vision System: Image Processing Over an Incomplete Color-Space
Alireza Haghshenas, Maziar Sharbafi, Danial Esmaeeli, Omid Bakhshandeh
IPCV 2010 [bib]

NLP Calendar

I'm the new lead of the unofficially official NLP Calendar: After 12 years of the awesome NLP calendar that Joel started, I'm continuing the legacy! Please feel free to propose new conferences and workshops to be added to the list.
For all publications and other details, please see my CV.