I am a third-year Ph.D. student at UC Davis, working on multimodal agents, self-evolving harness and video understanding. I am fortunate to be advised by Prof. Lifu Huang. Introduce myself! My research focuses on multimodal understanding and generation, encompassing image editing, image generation, and video understanding, to bridge perception and generation through unified multimodal frameworks. I also explore RL-trained and LLM/MLLM-based agents, applying agent frameworks to diverse real-world applications such as design document editing and medical reasoning. My broader interests include retrieval-augmented generation (RAG), interpretability, and the development of efficient, scalable architectures for multimodal intelligence. Now, I am especially interested in

Before joining UC Davis, I obtained my master’s degree from Duke University in 2022, majoring in Electronic and Computer Engineering. Before that, I obtained my bachelor’s degree from Southern University of Science and Technology in 2019, majoring in Financial Mathematics.

I am always open to collaboration in many domains. If you want to collaborate with me or have any interesting ideas or thoughts want to discuss with me, feel free to contact me: qzlin@ucdavis.edu and zihao.lin.0123@gmail.com.

Experiences

Research Scientist Intern @ Meta MRS Research Team
Host: Qifan Wang
June 2026 - Sep. 2026

Research Scientist Intern @ Adobe Research
Host: Wanrong Zhu
June 2025 - Nov. 2025

Research Scientist Intern @ Adobe Research
Host: Zichao (Jack) Wang
May 2024 - Nov. 2024

Research Intern @ IDEA Research
Host: Haiqin Yang Feb. 2023 - June 2023

News

• 09/08/2026: One paper about efficient RL for image generation has been accepted to AACL 2026.
• 08/20/2026: One paper about zero-shot 3D Visual Grounding has been accepted to EMNLP 2026.
• 08/20/2026: One paper about multi-layer multi-modal document editing has been accepted to TMLR 2026.
• 07/31/2026: Two papers about agentic video editing and game video glitch detection have been accepted to ACMMM 2026.
• 01/04/2026: One paper about conversational user simulation has been accepted to EACL 2026.
• 12/26/2025: I will join Meta as a Research Scientist Intern in June 2026.
• 11/03/2025: One paper about reasoning-based multi-layer design document editing accepted to NeurIPS 2025 Workshop Multimodal Algorithm Reasoning, awarded as a spotlight paper.
• 09/18/2025: Two papers about context compression and multi-modal interpretability accepted to NeurIPS 2025.
• 08/20/2025: One paper about reasoning-based image generation accepted to EMNLP 2025, awarded as a spotlight paper.
• 02/28/2025: One paper about question generation accepted to NAACL Demo Track 2025.
• 09/20/2024: Four papers about multi-modal agent, interleaved reasoning, and interpretability accepted to EMNLP 2024.
• 05/16/2024: One paper about memory editing accepted to ACL 2024.