Skip to main content

A world-class team pushing to advance the field of artificial intelligence and its incorporation into AI-Native products used by millions.

Research

We want our work to advance fundamental research in AI, NLP and machine learning and to shorten the time from discovery to deployment. In addition to scientific progress in its own right, research also informs every aspect of our product and how our millions of users interact with ASAPP products every day. Having a leading research team allows us to achieve significant commercial milestones while remaining directly engaged with the academic community as we endeavor to advance the field.

How We Research

We invest in long term, often counterintuitive and high-risk research projects. This work advances understanding in our fields of interest and are valuable to ASAPP’s bold vision of applying machine learning to historically complex problems.

  • Our research focuses on our ambitious long-term goals.This vision is guided both by ASAPP's specific product goals as well as the collective scientific vision of our research group
  • We treat long-term research projects as a series of smaller steps, each of which can be deployed in live production to give us better understanding of a problem and data for the next steps
  • Members of the research group remain academically active. This happens via the publication of scientific results, both positive and negative, and via ongoing and open collaboration with the academic community
  • Members are actively encouraged to pursue their own research interests independent of ASAPP's research goals, and to publish their work in leading AI, ML and NLP conferences.
  • A diversity of scientific backgrounds, ideas and opinions is critical to our successful research culture and drives the most innovative outcomes for our clients

Some Selected Publications

  • "Style Transfer from Non-Parallel Text by Cross-Alignment"
    Tianxiao Shen/MIT, Tao Lei/ASAPP, Regina Barzilay/MIT, Tommi Jaakkola/MIT
  • "Training RNNs as Fast as CNNs"
    Tao Lei/ASAPP, Yu Zhang/MIT, Yoav Artzi/ASAPP+Cornell University
  • "Rationalizing Neural Predictions"
    Tao Lei/MIT, Regina Barzilay/MIT, Tommi Jaakkola/MIT
  • "High-Risk Breast Lesions: A Machine Learning Model to Predict Pathologic Upgrade and Reduce Unnecessary Surgical Excision"
    Manisha Bahl/MGH, Regina Barzilay/MIT, Adam Yedidia/MIT, Nicholas Locascio/MIT, Lili Yu/MIT, Constance Lehman/MGH
  • "Learning High-Level Planning from Text"
    S.R.K. Branavan/MIT, Nate Kushman/MIT, Tao Lei/MIT, Regina Barzilay/MIT
  • "Learning to Win by Reading Manuals in a Monte-Carlo Framework"
    S.R.K. Branavan/MIT, David Silver/MIT, Regina Barzilay/MIT
  • "Reinforcement Learning for Mapping Instructions to Actions"
    S.R.K. Branavan/MIT, Harr Chen/MIT, Luke Zettlemoyer/MIT, Regina Barzilay/MIT

Academic Advisors

Prof. Regina Barzilay

MIT

Prof. Sam Bowman

NYU

Prof. Kyunghyun Cho

NYU

Prof. Alexander Rush

Harvard

Prof. Adler Perotte

Columbia

Prof. Luke Zettlemoyer

University of Washington

Research Team

Sam Altschul, PhD

University of Chicago

Prof. Yoav Artzi

Cornell Tech

S.R.K. Branavan, PhD

MIT

Howard Chen

Cornell Tech

Hui Dai, PhD

University of Chicago

Ethan Elenberg, PhD

University of Texas – Austin

Anna Folinsky, PhD

Caltech

Michael Griffiths

Skidmore

Shawn Henry, PhD

University of Chicago

Tao Lei, PhD

MIT

Nicholas Matthews

MIT

Karthik Narasimhan, PhD

MIT

Hugh Perkins

Cambridge

Max Sperlich

Georgetown

Prof. David Sontag

MIT

Sida Wang, PhD

Stanford

Jeremy Wohlwend

MIT

Lili Yu, PhD

MIT