Yibo Liu

CS PhD Candidate @ UVic

google-scholar
github
linkedin
twitter
CV

About

I am Yibo Liu, a second year Ph.D. student in Computer Science at University of Victoria (UVIC), working with Prof. Dr. Teseo Schneider . Prior to this, I obtained my M.S. in Computer Science from New York University (NYU) and B.Eng. in Electronic Engineering from Beijing University of Posts and Telecommunications (BUPT). Previously, I have interned at Microsoft Research Asia (MSRA).

My current research focuses on computer graphics, geometry modeling, and AI4Science, particularly on leveraging generative AI (e.g. LLMs) and learning-based paradigms to enhance physics-based simulations.

Experience

  • Univerisity of Victoria | Sept 2023 - Present
    PhD student in Computer Science
    Supervisor: Prof. Dr. Teseo Schneider

  • Geometric Computing Lab, New York University | 2022 - 2023
    Independent Study, supervised by Prof. Dr. Teseo Schneider and Prof. Dr. Daniele Panozzo
    Conducting research on GPU acceleration for contact simulations in PolyFEM library.

  • Microsoft Research Asia | Aug 2020 - Feb 2021
    R&D Intern at Data, Knowledge and Intelligence group, full-time, onsite
    Conducted research on recommending charts from tables (Table2Charts) using reinforcement learning, delivering the technique to Bing Search and Excel Spreadsheet Intelligence, and developed a multilingual key-phrase extraction algorithm used in Forms Ideas and Teams polls.

  • CILVR Lab, New York University | Mar 2020 - May 2021
    Independent Study, supervised by Prof. Dr. Iacer Calixto and Dr. Clara Vania
    Conducting research on learning robust mulilingual multimodal knowledge graph representations.

  • Center for Speech and Language Technologies, Tsinghua University | 2018 - 2019
    Research intern, supervised by Prof. Dr. Dong Wang
    Conducting research on poetry generation with rythm (YunLv) as soft constraint.

Publication

Neural Kinematic Bases for Fluids

Yibo Liu, Zhixin Fang, Sune Darkner, Noam Aigerman, Kenny Erleben, Paul Kry, Teseo Schneider

[Paper] [Demo]

Emergent Crowds Dynamics from Language-Driven Multi-Agent Interactions

In Submission

[Paper] [Demo]

MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI

Xiang Yue, Yuansheng Ni, Kai Zhang, Tianyu Zheng, Ruoqi Liu, Ge Zhang, Samuel Stevens, Dongfu Jiang, Weiming Ren, Yuxuan Sun, Cong Wei, Botao Yu, Ruibin Yuan, Renliang Sun, Ming Yin, Boyuan Zheng, Zhenzhu Yang, Yibo Liu, Wenhao Huang, Huan Sun, Yu Su, Wenhu Chen

[CVPR 2024 Best Paper Nomination] Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 9556-9567, Jun. 2024

[Paper] [Web Page]

Endowing Language Models with Multimodal Knowledge Graph Representations

Ningyuan Huang, Yash R. Deshpande, Yibo Liu, Houda Alberts, Kyunghyun Cho, Clara Vania, Iacer Calixto

arXiv 2206.13163, Jun. 2022

[Paper]

VisualSem: a high-quality knowledge graph for vision and language

Houda Alberts, Ningyuan Huang, Yash Deshpande, Yibo Liu, Kyunghyun Cho, Clara Vania, Iacer Calixto

Proceedings of the 1st Workshop on Multilingual Representation Learning, pp. 138-152, Nov. 2021. (colocated with EMNLP 2021)

[Paper]

Table2Charts: Recommending Charts by Learning Shared Table Representations

Mengyu Zhou, Qingtao Li, Xinyi He, Yuejiang Li , Yibo Liu, Wei Ji, Shi Han, Yining Chen, Daxin Jiang, Dongmei Zhang.

Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pp. 2389-2399, Aug. 2021

[Paper] [Code]

Open-Source Contribution

Neural Topic Model Library

  • The first Python library containing all cutting-edge neural topic models, collaborated with Leilan Zhang.
  • Refactored the codebase framework and rewrote the interfaces to provide APIs compatible with the Gensim LDA library.

[GitHub | 431 star ⭐️ | 80 fork]

PolyFEM

  • Developed GPU acceleration for contact simulations.
  • Profiled contact solvers in the PolyFEM library, identified performance bottlenecks.

[GitHub | 562 star ⭐️ | 88 fork]

Projects

→ Full list

Lock-free Linked List Library for GPUs

  • This project develops a CUDA-based singly linked list library, with all operations running on GPUs to fully exploit their parallelism.
  • It introduces the first lock-free linked list algorithm on CUDA, adapted from prior implementations on multi-core CPUs.

[GitHub] [Report]

Talks

Oct 2021 I presented our paper "VisualSem: a high-quality knowledge graph for vision and language" at EMNLP Workshop on Multilingual Representation Learning 2021.

Peer Review Service

AAAI Conference on Artificial Intelligence (AAAI) Reviewer, 2025

International Conference on Learning Representations (ICLR) Reviewer, 2025

Transactions on Visualization and Computer Graphics (TVCG) Reviewer, 2024

Volunteer

SIGGRAPH Asia 2024 Student Volunteer | Tokyo, Japan | Dec 2024

Teaching Assistantship

Introduction to Computer Graphics | University of Victoria | 2025 summer

Data Mining | University of Victoria | 2025 spring, 2024 winter, 2023 fall

Introduction to C++ | University of Victoria | 2024 fall

Software Architecture | University of Victoria | 2024 fall

Last Updated: 9/7/2025, 9:15:00 PM