Liwei Wang

I am an Assistant Professor in Computer Science and Engineering department at The Chinese University of Hong Kong (CUHK). Before coming to HK, I have worked for more than two years as a Senior Researcher in Tencent AI Lab at Bellevue, US.

I got my PhD from Computer Science Department, University of Illinois at Urbana-Champaign, advised by Prof. Svetlana Lazebnik. Here is my Short Bio.

The Language and Vision (LaVi) Lab, which I founded at the Department of Computer Science and Engineering at CUHK, conducts research in Natural Language Processing (NLP) and Computer Vision, with a particular emphasis on the intersection of vision and language. Our work encompasses a range of topics including Language+Vision, Large Language Models, Multi-modal Large Models, and Embodied AI. If you want to join LaVi Lab, please send an email to

Email / Google Scholar / Publications / Lab website (soon) /

  • 2024/06:
    We are hiring Researchers, Interns, Postdocs and Phd students to work on topics of Embodied AI, Vision+Language, LLMs, multi-modal LLMs, and diffusion models.
  • 2024/05:
    Will serve as an Area Chair of NeurIPS 2024.
  • 2023/10:
    Our work on LLMs Preference Model have been accepted to EMNLP 2023!
  • 2023/10:
    Our CLEVA, the comprehevsive Chinese Language Model Evalution Platform, has been accepted to EMNLP 2023 System Demonstration!
  • 2023/07:
    I was invited to give a talk in the IAS Workshop on Mathematical Theory for Emergent Intelligence.
  • 2023/07:
    Our work on Vision-Language Parameter-Efficient Tuning has been accepted to ICCV 2023.
  • 2023/07:
    Our work on reasoning with LLMs has been accepted to ACL 2023.
  • 2023/06:
    Will serve as an Area Chair of CVPR 2024
  • 2022/10:
    LaVi's two EMNLP 2022 Long papers on Dialogue research are accepted!
  • 2022/10:
    Serving as an Area Chair of CVPR 2023
  • 2022/07:
    Our LaVi team won the annual NLP challenge LIC 2022 (Multi-modal Video Understanding track) hosted by CCF and CIPS, check department news here .
  • 2022/07:
    One ECCV paper is accepted
  • 2022/03:
    Three papers got accepted to CVPR 2022 including our new work on Language + Vision.
  • 2022/02:
    One ACL 2022 long paper from our group on "Probing Pre-trained Models" has bee accepted.
  • 2022/02:
    Serving as an Area Chair of ECCV 2022 .
Recent Research Highlights

My students or interns are indicated by '*'. Click full publication list

Making Long-Context Language Models Better Multi-Hop Reasoners

Yanyang Li*, Shuo Liang*, Michael R. Lyu, Liwei Wang
ACL 2024

Beyond Embeddings: The Promise of Visual Table in Multi-Modal Models

Yiwu Zhong*, Ziyuan Hu*, Michael R. Lyu, Liwei Wang
Arxiv 2024

Towards Learning a Generalist Model for Embodied Navigation

Duo Zheng*, Shijia Huang*, Lin Zhao, Yiwu Zhong, Liwei Wang
CVPR 2024     Code

Learning Preference Model for LLMs via Automatic Preference Data Generation

Shijia Huang*, Jianqiao Zhao*, Yanyang Li*, Liwei Wang
EMNLP 2023 Long Paper

CLEVA: Chinese Language Models EVAluation Platform

Yanyang Li, Jianqiao Zhao, Duo Zheng, Zi-Yuan Hu, Zhi Chen, Xiaohui Su, Yongfeng Huang, Shijia Huang, Dahua Lin, Michael R. Lyu, Liwei Wang
EMNLP 2023 System Demonstration     Project

VL-PET: Vision-and-Language Parameter-Efficient Tuning via Granularity Control

Ziyuan Hu*, Yanyang Li*, Michael R. Lyu, Liwei Wang
ICCV, 2023     Code

Multi-View Transformer for 3D Visual Grounding

Shijia Huang*, Yilun Chen, Jiaya Jia, Liwei Wang
CVPR, 2022     Code

Eliciting Knowledge from Large Pre-Trained Models for Unsupervised Knowledge-Grounded Conversation

Yanyang Li*, Jianqiao Zhao*, Michael R. Lyu, Liwei Wang
EMNLP, 2022 Long Paper, Code

FlowEval: A Consensus-Based Dialogue Evaluation Framework Using Segment Act Flows

Jianqiao Zhao*, Yanyang Li*, Wanyu Du*, Yangfeng Ji, Dong Yu, Michael R. Lyu, Liwei Wang
EMNLP, 2022 Long Paper, Code and Dataset

Probing Structured Pruning on Multilingual Pre-trained Models: Settings, Algorithms, and Efficiency

Yanyang Li*, Fuli Luo, Runxin Xu, Songfang Huang, Fei Huang, Liwei Wang
ACL, 2022, long paper    

SAT: 2D Semantics Assisted Training for 3D Visual Grounding

Zhengyuan Yang, Songyang Zhang, Liwei Wang, Jiebo Luo
ICCV, 2021, Oral Presentation     Code

Improving Weakly Supervised Visual Grounding by Contrastive Knowledge Distillation

Liwei Wang, Jing Huang, Yin Li, Kun Xu, Zhengyuan Yang, Dong Yu
CVPR, 2021     Code

Robust Dialogue Utterance Rewriting as Sequence Tagging

Jie Hao, Linfeng Song, Liwei Wang, Kun Xu, Zhaopeng Tu, Dong Yu
EMNLP, 2021,     Code

Comprehensive Image Captioning via Scene Graph Decomposition

Yiwu Zhong*, Liwei Wang, Jianshu Chen, Dong Yu, Yin Li
ECCV, 2020     Code

Improving One-stage Visual Grounding by Recursive Sub-query Construction

Zhengyuan Yang, Tianlang Chen, Liwei Wang, Jiebo Luo
ECCV, 2020     Code

MART: Memory-Augmented Recurrent Transformer for Coherent Video Paragraph Captioning

Jie Lei*, Liwei Wang, Yelong Shen, Dong Yu, Tamara Berg, Mohit Bansal
ACL, 2020     Code

A Fast and Accurate One-Stage Approach to Visual Grounding​

Zhengyuan Yang*, Boqing Gong, Liwei Wang, Wenbing Huang, Dong Yu, Jiebo Luo
ICCV, 2019, Oral Presentation     Code

Fast, Diverse and Accurate Image Captioning Guided By Part-of-Speech

Aditya Deshpande, Jyoti Aneja, Liwei Wang, Alexander Schwing, D. A. Forsyth
CVPR, 2019, Oral Presentation

Learning Two-Branch Neural Networks for Image-Text Matching Tasks

Liwei Wang, Yin Li, Jing Huang, Svetlana Lazebnik
TPAMI, 2018     Code

Learning structural motif representations for efficient protein structure search

Yang Liu, Qing Ye, Liwei Wang, Jian Peng
Bioinformatics, 2018     Code

Diverse and Accurate Image Description Using a Variational Auto-Encoder with an Additive Gaussian Encoding Space

Liwei Wang, Alex Schwing, Svetlana Lazebnik
NeurIPS, 2017

Flickr30k Entities: Collecting Region-to-Phrase Correspondences for Richer Image-to-Sentence Models

Bryan Plummer, Liwei Wang, Chris M. Cervantes, Juan C. Caicedo, Julia Hockenmaier, Svetlana Lazebnik
IJCV, 2016     Project

Learning Deep Structure-Preserving Image-Text Embeddings

Liwei Wang, Yin Li, Svetlana Lazebnik
CVPR, 2016     Code

(Templates adapted from Arun's webpage.)