Lost in the Tail: Addressing Geographic Imbalance in Urban Visual Place Recognition
Zhiyao Shu, Jiacheng Yang, Yang Lu, Waishan Qiu, Chuan Li, Da Chen
I am a Ph.D. student in Information Science & Technology at George Mason University.
I work on computer vision and human–AI collaboration, building interactive systems that help autonomous agents understand and reason about what they sense and respond reliably in messy real-world scenarios.
🎉 Our paper “Lost in the Tail: Addressing Geographic Imbalance in Urban Visual Place Recognition” was accepted to ECCV 2026.
Began my PhD in Information Science & Technology at George Mason University.
Released MindCoder v2 — generate full reports in one click. Try it at mindcoder.ai.
Zhiyao Shu, Jiacheng Yang, Yang Lu, Waishan Qiu, Chuan Li, Da Chen
Jie Gao, Alok Prakash, Zhiyao Shu, Shun Yi Yeo, Chien-Ming Huang, Ziang Xiao, Mark Dredze
Zhiyao Shu, Xiangguo Sun, Hong Cheng
A Bayesian framework that infers the intended plan of a person from multimodal cues — speech, pointing gestures, and facial expressions — so robots can interpret and act on human intent during collaborative interaction.
A full-stack tool for flexible, structured inductive qualitative analysis. Automates open coding, axial coding, and concept development with GPT-5, producing one-click reports to support insight presentation.
A couple hundred new arXiv papers land in my categories every week, and maybe a dozen are worth my time. So I built an agent to find them — it scores each paper against my actual research and posts the 20 best to my Discord every Monday. Serverless, and free.
A summer road trip from Seattle to Fairfax: wild horses over the Columbia, Devils Tower, the Badlands, a modernist pilgrimage in Columbus, Indiana, and Fallingwater on the way home. With a clickable route map.
Open to research collaborations.