Zihao (Zack) Zhu

I am an incoming Ph.D. student in Electrical and Computer Engineering at Texas A&M University, co-advised by Prof. Xiaoning Qian and Prof. Zhiwen Fan. I received my B.S. in Computer Science from Texas A&M University, where I worked with Prof. Zhengzhong Tu.

My research aims to develop trustworthy and uncertainty-aware AI systems that can perceive, reconstruct, and reason about the visual world, with applications in 3D reconstruction, image restoration, generation, and multimodal understanding.

Email / CV / Scholar / GitHub

News

Jun	2026	I will join TAMU ECE as a Ph.D. student in Fall 2026.
May	2026	Trust3R is accepted to ICML 2026.
May	2026	We released 4KLSDB, a large-scale native-4K dataset for image restoration and generation (CVPR 2026 DataCV Workshop).
Oct	2025	We won 3rd place in the COGS Challenge for compact 3D representation at ICCV 2025.

Publications

Representative papers are highlighted. ^* denotes equal contribution and ^† denotes the corresponding author.

	Trust It or Not: Evidential Uncertainty for Feed-Forward 3D Reconstruction with Trust3R Zihao Zhu^, Wenyuan Zhao^, Nuo Chen, Chao Tian^†, Zhiwen Fan^† ICML, 2026 arXiv / project page / code / bibtex Equips feed-forward 3D reconstruction with probabilistically grounded, evidential pointmap uncertainty, so you know which parts of the predicted geometry to trust.
	4KLSDB: A Large-Scale Dataset for 4K Image Restoration and Generation Zihao Zhu, Kuan-Ru Huang, Zhaoming Xu, Renjie Li, Bo Wu, Ruizheng Bai, Mingyang Wu, Sayak Paul, Zhengzhong Tu^† CVPR Workshops (DataCV), 2026 arXiv / project page / code & data / bibtex 129,484 carefully curated native-4K images for training and benchmarking super-resolution and text-to-image diffusion models.
	How Independent are Large Language Models? A Statistical Framework for Auditing Behavioral Entanglement and Reweighting Verifier Ensembles Chenchen Kuai^, Jiwan Jiang^, Zihao Zhu, Hao Wang, Keshu Wu, Zihao Li, Yunlong Zhang, Chenxi Liu, Zhengzhong Tu, Zhiwen Fan, Yang Zhou^† arXiv preprint, 2026 arXiv A statistical framework for auditing hidden behavioral dependencies among LLMs and reweighting verifier ensembles accordingly.
	Agent Banana: High-Fidelity Image Editing with Agentic Thinking and Tooling Ruijie Ye^, Jiayi Zhang^, Zhuoxin Liu^, Zihao Zhu, Siyuan Yang, Li Li, Tianfu Fu, Franck Dernoncourt, Yue Zhao, Jiacheng Zhu, Ryan Rossi, Wenhao Chai, Zhengzhong Tu^† arXiv preprint*, 2026 arXiv / project page An agentic framework for multi-turn, high-fidelity editing of ultra-high-resolution images via context folding and layer decomposition.
	HeadsUp! High-Fidelity Portrait Image Super-Resolution Renjie Li, Zihao Zhu, Xiaoyu Wang, Zhengzhong Tu^† arXiv preprint, 2025 arXiv A single-step diffusion model that restores and upscales portrait images end-to-end, with the new PortraitSR-4K dataset.

Projects

InstantSplat++
Open-source project, PHAI Lab, 2026
code

An improved InstantSplat framework for sparse-view, large-scale scene reconstruction with Gaussian Splatting.

Service

Journal Reviewer: IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)

Template from Jon Barron's website.