PhD Student in Artificial Intelligence

Qi Xu 徐淇

I am a PhD student at Westlake University, advised by Prof. Peidong Liu. My research focuses on embodied AI, multimodal large language models, and 3D Vision.

I am interested in building intelligent systems that perceive, reconstruct, reason about, and act within physical environments from visual and multimodal observations. I previously received my master's and bachelor's degrees from Wuhan University.

Email GitHub CV Google Scholar

Education & Experience

2026 - Present

Westlake University

PhD Student in Artificial Intelligence, supervised by Prof. Peidong Liu.

Research: Embodied AI, Multimodal Large Language Models.

2025 - 2026

ByteDance, TikTok Live

Research Intern, LLM Applications Team. Mentored by Xiangtai Li. Worked on multimodal model research and large-scale training systems.

2023 - 2026

Wuhan University

Master in Pattern Recognition and Intelligent Systems, supervised by Prof. Shunping Ji.

Research: Computer Vision, 3D Reconstruction, Multimodal Learning.

2019 - 2023

Wuhan University

B.S. in Spatial Information and Digital Technology.

News

2026

HiCI was accepted to ICML 2026.

2026

Towards One-to-Many Temporal Grounding was accepted to ICML 2026.

2025

SIU3R received a NeurIPS 2025 Spotlight.

Research Interests

Embodied AI

Perception, memory, and spatial reasoning foundations for agents that interact with physical environments.

Multimodal LLMs

Spatial-temporal reasoning over images, videos, language, and 3D observations for grounded multimodal intelligence.

3D Vision

Scene reconstruction, generation, and geometric representations from sparse, unposed, or multimodal observations.

Selected Publications

Qi Xu is underlined. * denotes equal contribution.

NeurIPS 2025 Spotlight

SIU3R: Simultaneous Scene Understanding and 3D Reconstruction Beyond Feature Alignment

Qi Xu*, Dongxu Wei*, Lingzhe Zhao, Wenpu Li, Zhangchi Huang, Shunping Ji, Peidong Liu

Alignment-free framework for state-of-the-art 3D reconstruction and scene understanding from unposed images via a shared pixel-aligned 3D representation.

Project ArXiv Code

ICML 2026

Towards One-to-Many Temporal Grounding

Qi Xu*, Yue Tan*, Shihao Chen, Jiahao Meng, Anran Wang, Shunping Ji, Hao Fei, Xiangtai Li

Introduces a systematic solution for one-to-many temporal grounding with a 56k-sample dataset and reinforcement learning using CoT-based rewards.

Project

arXiv 2026

Any 3D Scene is Worth 1K Tokens: 3D-Grounded Representation for Scene Generation at Scale

Dongxu Wei*, Qi Xu*, Zhiqi Li, Hangning Zhou, Cong Qiu, Hailong Qin, Mu Yang, Zhaopeng Cui, Peidong Liu

Performs 3D scene generation directly within an implicit 3D latent space using 3D Diffusion Transformers.

Project ArXiv Code

ICML 2026

HiCI: Hierarchical Construction-Integration for Long-Context Attention

Xiangyu Zeng, Qi Xu, Yunke Wang, Chang Xu

Efficient long-context modeling with hierarchical attention, extending LLaMA-2 to 100K tokens with only 5.5% additional parameters.

ArXiv Code

arXiv 2026

VideoZeroBench: Probing the Limits of Video MLLMs with Spatio-Temporal Evidence Verification

Jiahao Meng, Yue Tan, Qi Xu, Haochen Wang, Zhongwei Ren, Weisong Liu, Yuhao Wang, Renrui Zhang, Yunhai Tong, Haodong Duan

A hierarchical benchmark exposing the gap between answer correctness and genuine evidence-based reasoning in long-video understanding.

Project ArXiv Code

NeurIPS 2025

E-MoFlow: Learning Egomotion and Optical Flow from Event Data via Implicit Regularization

Wenpu Li, Bangyan Liao, Yi Zhou, Qi Xu, Pian Wan, Peidong Liu

Unsupervised framework for joint ego-motion and optical flow estimation using implicit neural representations and geometric constraints.

Project ArXiv Code

Contact

I am open to research conversations and collaborations on embodied AI, multimodal reasoning, 3D Vision, and vision-language systems that connect perception with action.

insomniaaac@qq.com github.com/insomniaaac Google Scholar