Hello! I am a Postdoctoral Young Investigator at AI2 on the PRIOR team, based in Seattle, WA.

My academic interests lie in computer vision, machine learning and their applications to real world problems. Specifically, I focus on multimodal representation learning, especially for high-level video understanding and reasoning. Prior to joining AI2, I received my Ph.D. in Computer Science and Engineering from Seoul National University where I was advised by Prof. Gunhee Kim.

Education

work Allen Institute for AI, Seattle, WA, USA ^{2023.01 - Current}

• Postdoctoral Young Investigator on the PRIOR team

school Seoul National University (SNU), Seoul, South Korea ^{2017.03 - 2023.02}

• Ph.D. in Computer Science and Engineering
• Advisor: Prof. Gunhee Kim
• Thesis: Improving Efficiency in Large-Scale Self-Supervised Video Representation Learning

school Seoul National University (SNU), Seoul, South Korea ^{2010.03 - 2017.02}

• B.S. in Computer Science and Engineering (minor in Statistics)

Publications

• ACAV100M: Automatic Curation of Large-Scale Datasets for Audio-Visual Video Representation Learning
  Sangho Lee*, Jiwan Chung*, Youngjae Yu, Gunhee Kim, Thomas Breuel, Gal Chechik and Yale Song (*: equal contribution)
  ICCV 2021
  CVPR 2021: The Third Workshop on Learning from Unlabeled Videos
  [paper] [code&dataset]

• Unsupervised Representation Learning via Neural Activation Coding
  Yookoon Park, Sangho Lee, Gunhee Kim and David Blei
  ICML 2021 (long talk)
  [paper] [poster] [slides] [code]

• Parameter Efficient Multimodal Transformers for Video Representation Learning
  Sangho Lee, Youngjae Yu, Gunhee Kim, Thomas Breuel, Jan Kautz and Yale Song
  ICLR 2021
  CVPR 2021: The Second Intertional Workshop on Large Scale Holistic Video Understanding
  [paper] [poster] [slides]

• Self-Supervised Learning of Compressed Video Representations
  Youngjae Yu*, Sangho Lee*, Gunhee Kim and Yale Song (*: equal contribution)
  ICLR 2021
  [paper] [poster] [slides]

• A Memory Network Approach for Story-based Temporal Summarization of 360° Videos
  Sangho Lee, Jinyoung Sung, Youngjae Yu and Gunhee Kim
  CVPR 2018
  ECCV 2018 Workshop on 360° Perception and Interaction
  [paper] [project page (poster/slides/bibtex)]

• A Deep Ranking Model for Spatio-temporal Highlight Detection from a 360° Video
  Youngjae Yu, Sangho Lee, Joonil Na, Jaeyoun Kang and Gunhee Kim
  AAAI 2018 (spotlight)
  [paper] [poster] [slides] [bibtex]

• A Read-Write Memory Network for Movie Story Understanding
  Seil Na, Sangho Lee, Jisung Kim and Gunhee Kim
  ICCV 2017
  ICCV 2017: The Joint Video and Language Understanding Workshop
  [paper] [code] [poster] [bibtex]

• Encoding Video and Label Priors for Multi-label Video Classification on YouTube-8M dataset
  Seil Na, Youngjae Yu, Sangho Lee, Jisung Kim and Gunhee Kim
  CVPR 2017 Workshop on YouTube-8M Large-Scale Video Understanding
  [paper] [code] [bibtex]

Work Experience

• PRIOR @ Allen Institute for AI Research Intern (March-June, August-December 2022)

Awards

• Excellent Ph.D. Thesis Award (SNU CSE, 2023)
  • Selected as the best doctoral thesis by Department of Computer Science and Engineering, Seoul National University
• Naver Ph.D. Fellowship (Naver, 2021)
  • Awarded to outstanding graduate students in the field of Computer Science for their exceptional academic research
• Youlchon AI Star Fellowship (Youlchon Foundation, Nongshim Group, 2021)
  • An award for those who made distinguished research achievements in core AI fields
• MovieQA Challenge @ ICCV 2017 Workshop (ICCV Workshop, 2017)
  • ICCV 2017 Workshop on the Joining Video and Language Understanding
  • Ranked 2nd place
• Google Cloud & YouTube-8M Video Understanding Challenge (CVPR Workshop, 2017)
  • CVPR 2017 Workshop on YouTube-8M Large-Scale Video Understanding
  • Ranked 8th place out of 655 teams (Top 2%)