Scholar
Jungsoo Lee
Google Scholar ID: tCxDlbwAAAAJ
Qualcomm AI Research
On-device AI
Robustness
Multi-modal learning
Follow
Homepage
↗
Google Scholar
↗
Citations & Impact
All-time
Citations
633
H-index
9
i10-index
9
Publications
16
Co-authors
11
list available
Contact
CV
Open ↗
GitHub
Open ↗
LinkedIn
Open ↗
Publications
6 items
ForeSea: AI Forensic Search with Multi-modal Queries for Video Surveillance
2026
Cited
0
FLoC: Facility Location-Based Efficient Visual Token Compression for Long Video Understanding
2025
Cited
0
Generalized Contrastive Learning for Universal Multimodal Retrieval
2025
Cited
0
Personalized OVSS: Understanding Personal Concept in Open-Vocabulary Semantic Segmentation
2025
Cited
0
Concept-Aware LoRA for Domain-Aligned Segmentation Dataset Generation
2025
Cited
0
CustomKD: Customizing Large Vision Foundation for Edge Model Improvement via Knowledge Distillation
2025
Cited
0
Resume (English only)
Academic Achievements
Paper 'GCL' accepted at NeurIPS 2025
Paper 'PerSeg' accepted at ICCV 2025
Paper 'CustomKD' accepted at CVPR 2025
Paper 'Wisdom of Crowds' accepted at ICCV 2023
Paper 'CAFA' accepted at ICCV 2023
Paper 'ReLD' accepted at ECML/PKDD 2023
Paper 'EcoTTA' accepted at CVPR 2023
Paper 'BiasEnsemble' accepted at AAAI 2023 as an oral presentation
Education
PhD from Korea Advanced Institute of Science and Technology (KAIST), advised by Professor Jaegul Choo.
Background
Currently a Senior Research Engineer at Qualcomm Korea. Main research interest is in multimodal learning.
Co-authors
11 total
Jaegul Choo
Associate Professor, Graduate School of Artificial Intelligence, KAIST
Sungha Choi
Qualcomm AI Research
Sanghun Jung
PhD Student in CSE at University of Washington
Co-author 4
Chaeyeon Chung
NVIDIA
Sunghyun Park
Qualcomm AI Research
Taewook Kim
Northwestern University
Jeonghoon Park
Korea Advanced Institute of Science and Technology (KAIST)
×
Welcome back
Sign in to Agora
Welcome back! Please sign in to continue.
Email address
Password
Forgot password?
Continue
Do not have an account?
Sign up