|
I teach machines to see like people and interact with people. As modern machines struggle to fully conceptualize the visual world, my research bootstraps machine learning using frameworks from behavioral and social sciences.
Bio: Ranjay Krishna is an Assistant Professor at the Allen School of Computer Science & Engineering. He co-directs the RAIVN lab at UW and is a member of technical staff at Microsoft SuperIntelligence. Prior to this, he directed the multimodal and embodied AI team at the Allen Institute. His work has been recognized with an NSF Career Award '26 and as one of MIT Technology Review's 35 under 35 Asia Pacific '25. He holds a bachelor's degree in Electrical & Computer Engineering and in Computer Science from Cornell University, a master's degree in Computer Science from Stanford University and a Ph.D. in Computer Science from Stanford University. His research lies at the intersection of computer vision, natural language processing, robotics, and human computer interaction. This research has received best paper honorable mentions at CVPR'25 and CSCW'23, best paper award nominee at CVPR'26, outstanding paper at NeurIPS'21 and ACL'21, and dozens of orals at CVPR, ACL, CSCW, NeurIPS, UIST, ICCV and ECCV, and has been reported by Science, Forbes, the Wall Street Journal, and PBS NOVA. His research has been supported by Google, Apple, Ai2, Amazon, SRC, Sony, Samsung, Cisco, Toyota Motor Inc, Toyota Research Institute, NSF, ONR, and Yahoo. RECENT PAPER HIGHLIGHTS
[Jul 2026] Our DiScoFormer paper received Spotlight award at ICML 2026, awarded to top 2.2% of submissions. [Jun 2026] Our Molmo2 paper was nominated for Best Paper award at CVPR 2026, awarded to top 74 submissions. [Jun 2026] Our Agile Deliberation paper received Highlight award at CVPR 2026, awarded to top 5% of submissions. [Jun 2026] Our VideoNet paper received Highlight award at CVPR 2026, awarded to top 5% of submissions. [Oct 2025] Our Latte paper received Oral award at EMNLP 2025, awarded to top 1.5% of submissions. [Oct 2025] Our TrajVIT paper received Highlight award at ICCV 2025, awarded to top 5% of submissions. [Jun 2025] Our Molmo paper received Best Paper Honorable Mention at CVPR 2025. [Jun 2025] Our Molmo paper will appear as a Oral at CVPR 2025, awarded to top 0.7% of submissions. [Apr 2025] Our interleaved scene graph paper will appear as a Spotlight at ICLR 2025, awarded to top 5% of submissions RECENT TALKS [July 2026] Keynote talk at WWW 2026 workshop on Agents for Recommendations & Online Marketplaces [Jun 2026] Public talk at Technology Alliance Discovery Series titled "What does it mean for a machine to understand the world through sight?" [Jun 2026] Keynote talk at CVPR'26 workshop on Multi-Modal Reasoning for AI Agents titled "It is Time to Rethink Grounding" [Jun 2026] Keynote talk at CVPR'26 workshop on Multimodal Learning and Applications titled "One Model Enabling Robotics, Computer Use and Motion Modeling" [Jun 2026] Keynote talk at CVPR'26 workshop on Visual Concepts titled "Why does it still suck? What is going on inside a vision-language model?" [Jun 2026] Keynote talk at CVPR'26 workshop on Unified Robotic Vision with Cross-Modal Sensing and Alignment titled "Reasoning models for Robotics" [Jun 2026] Keynote talk at CVPR'26 workshop on Embodied Reasoning in Action titled "Making Open Robotics Reasoning Models" [Jun 2026] Keynote talk at CVPR'26 workshop on Evaluation of Generative Foundation Model titled "Updating generative evaluations to meet modern demands" [Jun 2026] Keynote talk at CVPR'26 workshop on 3D-LLM/VLA titled "Zero-shot Sim2Real is possible for Manipulaton (Introducing MolmoSpaces & MolmoBot)" [Jun 2026] Keynote talk at CVPR'26 workshop on Test-time Scaling for Computer Vision titled "Visual Reasoning by Simulating Future Steps" [Jun 2026] Keynote talk at CVPR'26 workshop on Transformers for Vision titled "Architectures for Grounded Visual Reasoning (Introducing MolmoPoint & MolmoMotion)" [Jun 2026] Keynote at CVPR'26 workshop on Emerging Directions in Data for Multimodal Foundation Models titled"Truly Open, Reproducible Video-Language Models" [Jun 2026] Keynote talk at CVPR'26 workshop on AI for Visual Arts titled "User Controllability through Models, Interfaces Interventions" [Jun 2026] Keynote at CVPR'26 workshop on Multimodal Spatial Intelligence titled "Visual Reasoning is not working and we don’t know why" [Jun 2026] Keynote at CVPR'26 workshop on Agentic AI for Visual Media titled "From Vision to Action: Extracting Structure and Agency from Flat Pixels" [Jun 2026] Keynote at CVPR'26 workshop on Multimodal Alignment for a Pluralistic Society titled "Multilingual Pluralism through Dataset Interventions" ACademic Publications
2025
2024
2023
2022
EARLIER PUBLICATIONS
Non-Archival PAPERS
|
Assistant Professor
Computer Science & Engineering University of Washington Co-director of RAIVN Lab Member of Technical Staff @ Microsoft SuperIntelligence PAST AFFILIATIONS
Director of multimodal & embodied AI @ Allen Institute for Artificial Intelligence Visiting Faculty @ Google Research Visiting Faculty @ Apple Research Ph.D. @ Stanford University, 2021
Co-advised by Fei-Fei Li and Michael Bernstein. Curriculum Vitae [2026] Google scholar Research statement [2021] Teaching statement [2021] Diversity statement [2021] CONTACT
ranjay [at] cs [dot] washington [dot] edu Bill & Melinda Gates Center Room 304 3800 E Stevens Way NE, Seattle, WA 98195 TEACHING
University of Washington: CSE 599H: AI vs IA [2023] CSE 493G1: Deep learning [2025] [2024] [2023] CSE 455: Computer Vision [2025] [2024] Stanford University: CS231N: Convolutional Neural Networks for Visual Recognition [2021] [2020] CS131 Computer Vision: Foundations and Applications [2019] [2018] [2017] [crowdsourced class notes] RESEARCH GROUPPostDocs
Zhongzheng (Jason) Ren
(2025-) 
Jaemin Cho
(2025-) PhD students
Jieyu Zhang
(2020-) 
Benlin Liu
(2021-) 
Xiang Fan
(2023-) 
Linjie Li with Yejin Choi
(2023-) 
Chenhao Zheng
(2024-) 
Arjun Guru
(2025-) Long term collaborating PhD students
Arijit Ray with Kate Saenko
Former PostDocs
Wei-Chiu Ma (2023-2024)
Assistant Professor @ Cornell Former PhD students
M. Saygin Seyfioglu (2020-2025)
Applied Scientist @ Amazon 
Jae Sung Park (2020-2026)
Research Scientist @ Allen Institute Selected TalksVenue: CVPR 2024 - Computer Vision and Pattern Recognition
Panel: CVPR: past, present, and future Venue: CVPR 2020 - Computer Vision and Pattern Recognition
Title: Compositionally in Computer Vision [slides][video][workshop] Venue: CVPR 2020 - Computer Vision and Pattern Recognition
Title: Dense Captioning Events in Videos [slides][video][workshop] Venue: ECCV 2016 - European Conference on Computer Vision
Title: Visual Relationship Detection with Language Priors [pdf][project][slides][poster][video] Venue: CHI 2016 - Conference on Human Factors in Computer Systems
Title: Embracing Error to Enable Rapid Crowdsourcing [pdf][slides] MISCELLANEOUS
Trailer for a documentary
Venue: PBS NOVA Title: Can we build a brain? Year: 2018 Complete documentary
Venue: PBS NOVA Title: Can we build a brain? Year: 2018 |
