Scholar
Matthew Gwilliam
Google Scholar ID: lB9WkQ0AAAAJ
Research Scientist, TikTok
Citations & Impact
Citations
545
H-index
8
i10-index
8
Publications
20
Co-authors
0
Publications
Resume (English only)
Academic Achievements
- Publications:
- 1. 'Evolutionary Caching to Accelerate Your Off-the-Shelf Diffusion Model', Under Review, 2025.
- 2. 'How to Design and Train Your Implicit Neural Representation for Video Compression', Under Review, 2025.
- 3. 'Accelerate High-Quality Diffusion Models with Inner Loop Feedback', Under Review, 2025.
Research Experience
- Organized the workshop on Implicit Neural Representation for Video at CVPR 2024 during his PhD. Interned at Amazon, SRI, and NVIDIA, working on applications including recommendation, video retrieval, and efficiency for diffusion image generation. Research areas covered a wide range, including video compression, efficient image generation, and recognition.
Education
- PhD: Department of Computer Science, University of Maryland (UMD), Advisor: Professor Abhinav Shrivastava.
Background
- Research Interests: Computer Vision, Unsupervised Learning, Multimodal Representation Learning. Brief Introduction: A research scientist at TikTok focusing on unsupervised learning methods for universal image representation models, especially those based on diffusion and implicit neural representations. Interested in video retrieval, compression, generation; image classification, clustering, etc.
Co-authors