ClawPet
A voice-first embodied AI companion with simple physical expression, agent-to-agent chat, and a 3D-printable shell for desktop life.
Tong Zhang Stone
I am a Ph.D. student in Computer Science at Fudan University, advised by Prof. Yun Peng and Prof. Tao Xie. I received my M.S. in Computer Engineering from the University of California, Irvine, where I worked with Prof. Xiaohui Xie, and my B.Eng. from the Southern University of Science and Technology, where I worked with Prof. Yu Zhang.
My research focuses on Coding Agent and Multimodal Reasoning. Feel free to get in touch!
I believe the most exciting frontiers for AI are curiosity, which enables efficient data intervention, and abstraction, which reveals the structure of knowledge.
“What magical trick makes us intelligent? The power of intelligence stems from our vast diversity, not from any single, perfect principle.”
Marvin Minsky, The Society of Mind
MockMyInterview
An AI-powered mock interview experience for practicing technical and behavioral interviews with interactive feedback.
LaTeX Blog Template
A reusable LaTeX blog system template with blog-style landing pages, automated compilation, and GitHub Pages deployment.
GEB-Bench: Abstract Structures Told in Many Voices
GEB-Bench reveals that models can recognize an abstract structure within one representation yet struggle to carry the same relation across scenes, stories, mathematics, and programs.
InfiniteWorld: Separating Visual Evidence from Familiar Physics
InfiniteWorld tests whether vision-language models follow observed videos when their physical rules depart from familiar Earth priors, revealing a persistent bias toward the world they already know.
When Does Recurrence Become an Algorithm? Convergence Selection in Weight-Tied Looped Transformers
This study shows how training loop count shapes a moving computation frontier in weight-tied Transformers, including when further recurrence completes a solution or destabilizes it.
Knowing When to Stop Thinking
EFEReasoning dissects how progress, disagreement, stopping, and online world-model updates affect test-time planning, showing that a better internal objective does not always produce a better action.
LiveUIBench: Evaluating Agent as App Workflows
LiveUIBench evaluates agent-as-app orchestrations across changing goals, revisions, service actions, saved state, and return visits through a shared runtime.
When LLM Defenses Backfire: Characterizing Safety, Performance, and Cost Trade-offs
This study shows how jailbreak defenses trade safety against task performance, over-refusal, and inference cost across open-source language models.
Beyond Pixels: Human-Readable SVG Generation
S²VG² uses vision-language models to generate scalable vector graphics whose structure remains readable and editable by people.
For a more complete publication list, please see my CV.
Tong Zhang, Zhiyuan Shi, Yun Peng, and Tao Xie, “GEB-Bench: Abstract Structures Told in Many Voices”, Under Review.
Tong Zhang, Yuxuan Chen, and Yun Peng, “InfiniteWorld: Separating Visual Evidence from Familiar Physics”, Under Review.
Tong Zhang, Zexin Li, Simin Chen, and Yun Peng, “When LLM Defenses Backfire: Characterizing Safety, Performance, and Cost Trade-offs”, Under Review.
Tong Zhang, Junhao Hu, Yun Peng, and Tao Xie, “When Does Recurrence Become an Algorithm? Convergence Selection in Weight-Tied Looped Transformers”, Under Review.
Tong Zhang, Zhiyuan Shi, Yun Peng, and Tao Xie, “Knowing When to Stop Thinking: A Dissection of Expected Free Energy as a Test-Time Compute Controller for Latent World-Model Planning”, Under Review.
Tong Zhang, Yuxuan Cheng, Ruimin Zhang, Yun Peng, and Tao Xie, “LiveUIBench: Evaluating Agent as App Workflows”, Under Review.
Sinin Zhang, Yunfei Xie, Yuxuan Cheng, Haoyu Zhang, and Tong Zhang, “PhysNote: Self-Knowledge Notes for Evolvable Physical Reasoning in Vision-Language Model”, ES-Reasoning @ ICLR, 2026.
Tong Zhang, Haoyang Liu, Peiyan Zhang, Yuxuan Cheng, and Haohan Wang, “Beyond Pixels: Exploring Human-Readable SVG Generation for Simple Images with Vision Language Models”, Preprint, 2024.
Haoyu Ma, Tong Zhang, Shanlin Sun, Xiangyi Yan, Kun Han, and Xiaohui Xie, “CVTHead: One-shot Controllable Head Avatar with Vertex-feature Transformer”, IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2024.
Bu Jin, Xinyu Liu, Yupeng Zheng, Pengfei Li, Hao Zhao, Tong Zhang, Yuhang Zheng, Guyue Zhou, and Jingjing Liu, “ADAPT: Action-aware Driving Caption Transformer”, IEEE International Conference on Robotics and Automation (ICRA), 2023.
Sep. 2025 to Dec. 2025
Research Intern
Developed a WeChat Mini Program agent based on vision-language models, and designed agent memory together with experience extraction mechanisms.
May 2025 to Aug. 2025
Research Intern
Proposed an energy-based attention refinement method for image editing, and explored iterative loops that generate, reflect, and reason for multi-round controllable optimization.
Nov. 2024 to Apr. 2025
Research Intern
Studied defense mechanisms for large language models, their side effects, and token compression for audio language models.
May 2022 to Sep. 2022
Lightweight OCR Models for OpenCV
Implemented PP-OCRv2 detection in OpenCV Zoo, high-level C++ APIs, and evaluation metrics for text detection.
Erhu
Horse Racing (赛马)
Singing, badly
Hey Jude