Christopher Clark
I am a research scientist with PRIOR team at the non-profit AI research institute the Allen Institute for AI. My general research interests are in unified vision and language systems and out-of-domain generalization. My recent projects have involved training models that can complete many multi-modal tasks with a shared architecture. Previously I have worked on training models to play the drawing and guessing game Iconary, and my PhD focused on ways to prevent models using spurious correlations or non-generalization patterns found in the training data.
I received my PhD from UW where I was advised by Luke Zettlemoyer. Before that I was a Predoctoral Young Investigator at the AI2 and completed a Masters at the University of Edinburgh.
Publications
A-OKVQA: A Benchmark for Visual Question Answering using World Knowledge
Dustin Schwenk, Apoorv Khandelwal, Christopher Clark, Kenneth Marino, Roozbeh Mottaghi
In EMNLP 2021
[paper]
[code]
[project page]
Webly Supervised Concept Expansion for General Purpose Vision Models
Amita Kamath*, Christopher Clark*, Tanmay Gupta*, Eric Kolve, Derek Hoiem, Aniruddha Kembhavi
In ECCV 2022
[paper]
[code]
[demo]
[project page]
Iconary: A Pictionary-Based Game for Testing Multimodal Communication with Drawings and Text
Christopher Clark, Jordi Salvador, Dustin Schwenk, Derrick Bonafilia, Mark Yatskar, Eric Kolve, Alvaro Herrasti, Jonghyun Choi, Sachin Mehta, Sam Skjonsberg, Carissa Schoenick, Aaron Sarnat, Hannaneh Hajishirzi, Aniruddha Kembhavi, Oren Etzioni, Ali Farhadi
In EMNLP 2021
[paper]
[code]
BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions
Christopher Clark, Kenton Lee, Ming-Wei Chang, Tom Kwiatkowski, Michael Collins, Kristina Toutanova
In NAACL 2019
[paper]
[dataset]
[leaderboard]