Scholar
Tyler Zhu
Google Scholar ID: -aoybA4AAAAJ
PhD student, Princeton University
computer vision
machine learning
Follow
Homepage
↗
Google Scholar
↗
Citations & Impact
All-time
Citations
2,305
H-index
2
i10-index
1
Publications
6
Co-authors
2
list available
Contact
CV
Open ↗
Twitter
Open ↗
GitHub
Open ↗
Publications
5 items
Video Models Reason Early: Exploiting Plan Commitment for Maze Solving
2026
Cited
0
Perception Test 2025: Challenge Summary and a Unified VQA Extension
arXiv.org · 2026
Cited
0
Dynamic Reflections: Probing Video Representations with Text Alignment
2025
Cited
0
Attention IoU: Examining Biases in CelebA using Attention Maps
2025
Cited
0
Unifying Specialized Visual Encoders for Video Language Models
2025
Cited
0
Resume (English only)
Academic Achievements
Recipient of the Princeton President's Fellowship.
Paper 'MERV: A Multi-Encoder Representation of Videos' accepted to ICML 2025; preprint on arXiv, models on Hugging Face.
Paper 'xT: Nested Tokenization for Larger Context in Large Images' accepted to ICML 2024; preprint on arXiv.
Paper 'PaReprop: Fast Reversible Transformers' accepted as a spotlight at Transformers for Vision Workshop @ CVPR 2023.
Co-organizing the Transformers for Vision Workshop @ CVPR 2024.
Co-authors
2 total
Dan Hendrycks
Director of the Center for AI Safety (advisor for xAI and Scale)
Karttikeya Mangalam
CEO & Founder, SigIQ.ai
×
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