Our scientists have made seminal contributions to the development of AI and its application in healthcare.
Advancing the frontiers of generative ai.
Generating SOAP Notes from Doctor-Patient Conversations Using Modular Summarization Techniques
Kundan Krishna, Sopan Khosla, Jeffrey Bigham, Zachary C. Lipton
Association for Computational Linguistics (ACL), 2021
Learning to Diagnose with LSTM Recurrent Neural Networks
Zachary C. Lipton, David C. Kale, Charles Elkan, Randall Wetzel
International Conference on Learning Representations (ICLR) 2016
The Belebele Benchmark: a Parallel Reading Comprehension Dataset in 122 Language Variants
Lucas Bandarkar, Davis Liang*, Benjamin Muller, Mikel Artetxe, Satya Narayan Shukla, Donald Husa, Naman Goyal, Abhinandan Krishnan, Luke Zettlemoyer, Madian Khabsa
Belebele sheds light on the capabilities of LLMs on both reading comprehension and long-tail multilingual understanding (arXiv whitepaper). *Davis Liang is affiliated with Abridge.
Leading science, laser-focused on impact
Our team is laying the foundations for the development, deployment, and continuous improvement of AI systems that reshape the practice of medicine, unburdening clinicians while respecting their judgment.
Developing AI responsibly and transparently
Our commitment to building clinical AI systems that providers and healthcare organizations can trust sets us apart.
Provenance tracking for accountability
Our software allows clinicians to validate each part of a generated note against the underlying patient conversation transcript and audio recording.
Guardrails for reliability
Our team is innovating techniques for preventing mistakes before they happen, catching them before they hit production, and learning rapidly from feedback whenever we fall short.
Control through intuitive UIs
We've developed workflows that keep clinicians in charge through user-centric interfaces. Our tools help clinicians review, fact-check, and revise AI-generated notes before signing off.
Continuous innovation up and down the stackContact us
Speech recognition tuned to handle medical conversations across 14+ languages, coping with cross-talk, background noise, and an evolving landscape of maladies, medications, and practice patterns.
Sophisticated summarization pipelines that intelligently combine massive proprietary datasets and novel methodology with judicious use of cutting-edge large language models. Innovative methods for tying every summary span back to substantiating evidence in the underlying conversation.
Analysis and action
After processing the conversations, what have we learned? Our research goes beyond note generation to reveal insights that can guide improvements in the administration of healthcare.
Join our mission-driven teamExplore roles
See real-world improvement daily
Our systems help clinicians nationwide. Receive continual feedback and observe how your work assists providers and patients.
Learn and grow from experienced colleagues
Our team includes veterans from Meta AI, Microsoft Research, Amazon AI, Carnegie Mellon University, and more. Collaborate with colleagues at the top of their fields.
Publish pioneering research
Our culture encourages publishing in top journals and conferences. We value scientific contributions.
Make a difference
Improve clinical workflows, increase access to care, reduce burnout, and more. Your work positively impacts patient outcomes.