Welcome! I’m Haosen.
I am a first-year Master’s student in Computer Science at Northwestern University. Currently, I am working with Prof. Manling Li at NU-MLL-Group, collaborating with the Stanford Vision and Learning Lab. Previously, I was a research intern at the Shanghai AI Lab. I earned my bachelor’s degree in Data Science and Technology from the Hong Kong University of Science and Technology, where I worked under the guidance of Prof. Chi-Keung Tang and Prof. Yu-Wing Tai.
My research interests span Multi-modalities Generative AI, 3D Vision, Embodied AI, and Efficient AI. My goal is to empower machines with the ability to extract meaningful patterns and relationships from both structured data (e.g. text, images, and video) and unstructured 3D geometric data. Additionally, I hope to enhance the interpretability and explainability of these models, advancing us toward the development of human-centered and physically-grounded general artificial intelligence.
I am actively seeking a PhD position beginning in Fall 2026. If our research interests align, please feel free to connect!
🔥 News
- 2025.02: 🎉🎉 One paper accepted by CVPR 2024!
- 2025.02: 🎉🎉 Will serve as ACMMM 2025 Reviewer.
- 2024.09: 🎉🎉 Will serve as ICLR 2025 Reviewer.
- 2024.07: 🎉🎉 Will join Shanghai Artificial Intelligence Laboratory as a research intern.
- 2024.07: 🎉🎉 Two papers accepted by ECCV 2024!
- 2024.06: 🎉🎉 Awarded “Kaggle Competitions Expert”.
- 2024.06: 🎉🎉 Honored the Dean List Award in Spring 2023-24.
- 2024.06: 🎉🎉 Received a Silver medal 🥈 in “Image Matching Challenge 2024 - Hexathlon” (CVPR’24 Workshop), ranked 28th/ 929. Our solution was released.
- 2023.11: 🎉🎉 Received a Silver medal 🥈 in “Google - Fast or Slow? Predict AI Model Runtime”, ranked 40th/ 616. Our solution was released.
📝 Publications
* indicates equal contribution
Re-thinking Temporal Search for Long-Form Video Understanding
Jinhui Ye*, Zihan Wang*, Haosen Sun, Keshigeyan Chandrasegaran, Zane Durante, Cristobal Eyzaguirre, Yonatan Bisk, Juan Carlos Niebles, Ehsan Adeli, Li Fei-Fei, Jiajun Wu, Manling Li
Conference on Computer Vision and Pattern Recognition (CVPR), 2025
[Project Page] [Paper] [Project Code]
- We introduce LongVideoHaystack (LV-Haystack), a 480-hour dataset for keyframe search in long videos, with 15,092 human-annotated instances (SOTA scores 2.1% Temporal F1).
- Our framework T* reframed temporal search as spatial search with adaptive zooming, boosting GPT-4o from 50.5% to 53.1% and LLaVA-OV from 56.5% to 62.4% on LongVideoBench XL.
Auto-DAS: Automated Proxy Discovery for Training-free Distillation-aware Architecture Search
Haosen Sun, Peijie Dong, Zimian Wei, Shitong Shao, Lujun Li
European Conference on Computer Vision (ECCV), 2024
- We present Auto-DAS, an automatic proxy discovery framework using an Evolutionary Algorithm (EA) for training-free Distillation-aware Architecture Search (DAS).
- Auto-DAS generalizes well to various architectures and search spaces (e.g. ResNet, ViT, NAS-Bench-101, and NAS-Bench-201), achieving state-of-the-art results in both ranking correlation and final searched accuracy.
Auto-GAS: Automated Proxy Discovery for Training-free Generative Architecture Search
Lujun Li, Haosen Sun, Shiwen Li, Peijie Dong, Qifeng Liu, Wei Xue, Yike Guo
European Conference on Computer Vision (ECCV), 2024
- We introduce Auto-GAS, the first training-free Generation Architecture Search (GAS) framework enabled by an auto-discovered proxy, which achieves competitive scores with 110× faster search than GAN Compression.
Inpaint4DNeRF: Promptable Spatio-Temporal NeRF Inpainting with Generative Diffusion Models
Han Jiang*, Haosen Sun*, Ruoxuan Li*, Yu-Wing Tai, Chi-Keung Tang
Arxiv Preprint (Submitted to CVPR’24), Dec 2023
[Project Page] [Paper] [Project Code]
- Inpaint4DNeRF can generate prompt-based objects guided by the seed images and their 3D proxies while preserving multiview consistency. Our generative baseline framework is general which can be readily extended to 4D dynamic NeRFs.
Registering Neural Radiance Fields as 3D Density Images
Han Jiang*, Ruoxuan Li*, Haosen Sun, Yu-Wing Tai, Chi-Keung Tang
Arxiv Preprint, May 2023
- We proposes a method to align and merge pre-trained NeRF models of partially overlapping 3D scenes using a generalized registration pipeline, incorporating key point detection, point set registration, and universal pre-trained descriptor networks with contrastive learning strategy.
Additional Publications
- Measuring road safety achievement based on EWM-GRA-SVD: A decision-making support system for APEC countries, Faan Chen*, Lin Shi*, Yaxin Li, Qilin Wang, Haosen Sun, Xinyu Tang, Jiacheng Zu, Zhenwei Sun, Knowledge-Based Systems
🎖 Honors and Awards
- 2020.09 - 2024.07 HKUST Admissions Scholarship (Kerry Holdings Limited Scholarship, HK$280,000)
- 2024.06 The Dean List Award, Top 10%
- 2024.06 Silver Medal in CVPR’24 Workshop (Image Matching Challenge 2024 - Hexathlon), ranked 28th/ 929
- 2023.11 Silver Medal in Kaggle Competition (Google - Fast or Slow? Predict AI Model Runtime), ranked 40th/ 616
- 2022.08 Nomination for the Mr. Armin and Mrs. Lillian Kitchell Undergraduate Research Award
- 2019.10 Bronze Medal and the First Prize in the 36th Chinese Physics Olympiad (CPHO), Top 0.1%
- 2019.07 the Third Prize in the 28th China National Biology Olympiad (CNBO), Top 5%
📖 Educations
- 2024.09 - 2026.06 (now), M.S. in Computer Science, Northwestern University, USA
- 2020.09 - 2024.07, BSc in Data Science and Technology, Hong Kong University of Science and Technology (HKUST), Hong Kong
💻 Internships
-
07/2024 – 09/2024, Shanghai Artificial Intelligence Laboratory, China.
Research Intern, working closely with Dr. Peng Ye.
-
10/2023 – 05/2024, Hong Kong Generative AI Research and Development Center (HKGAI), Hong Kong.
Research Intern, working closely with Dr. Lujun Li.