Biography
I am Yimu Wang, a second-year Ph.D. student at UWaterloo. I obtained my master’s degree under the supervision of Prof. Lijun Zhang in the LAMDA Group led by Prof. Zhihua Zhou at Nanjing University. I was honored to spend a wonderful RA time at Tsinghua University with Prof. Jingjing Liu and Prof. Yang Liu and amazing experiences at Tencent Lightspeed & Quantum Studios, Alibaba, Netease Games, and Megvii. I also had a wonderful journey at Derivatives-China as a quant research intern.
Interests
- Multi-modal Learning
- Point Cloud Understanding
Education
PhD in Computer Science, 2022 - Now
University of Waterloo
MSc in Computer Science and Technology, 2018 - 2021
Nanjing University
BSc in Software Engineering, 2014 - 2018
Northwestern Polytechnic University
Recent News
Selected Publications and Preprints
(2023).
Balance Act: Mitigating Hubness in Cross-Modal Retrieval with Query and Gallery Banks.
EMNLP.
(2023).
InvGC: Robust Cross-Modal Retrieval by Inverse Graph Convolution.
Findings of EMNLP.
(2023).
Video-Text Retrieval by Supervised Sparse Multi-Grained Learning.
Findings of EMNLP.
(2023).
Cooperation or Competition: Avoiding Player Domination for Multi-target Robustness by Adaptive Budgets.
IEEE/CVF Conference on Computer Vision and Pattern Recognition.
(2023).
Multimodal Federated Learning via Contrastive Representation Ensemble.
International Conference on Learning Representations.
(2023).
Video-Text Retrieval by Supervised Multi-Space Multi-Grained Alignment.
arXiv:2302.09473.
(2022).
Multi-View Fusion Transformer for Sensor-Based Human Activity Recognition.
arXiv:2202.12949.
(2021).
Deep Unified Cross-Modality Hashing by Pairwise Data Alignment.
International Joint Conference on Artificial Intelligence.
(2021).
Classification of neurofibromatosis-related dystrophic or nondystrophic scoliosis based on image features using Bilateral CNN.
Med Phys.
(2020).
Piecewise Hashing: A Deep Hashing Method for Large-Scale Fine-Grained Search.
Pattern Recognition and Computer Vision.
(2020).
Searching Privately by Imperceptible Lying: A Novel Private Hashing Method with Differential Privacy.
ACM International Conference on Multimedia.
(2020).
Nearly Optimal Regret for Stochastic Linear Bandits with Heavy-Tailed Payoffs.
International Joint Conference on Artificial Intelligence.
(2020).
An Adversarial Domain Adaptation Network For Cross-Domain Fine-Grained Recognition.
IEEE Winter Conference on Applications of Computer Vision (WACV).
Experiences
Research Assitant
Researching on multi-modal learning (Retrieval) and Federated Learning under the supervision of Jingjing Liu and Yang Liu. We propose a framework that is able to distill knowledge from private users’ unimodal data and then fuse that knowledge into the multi-modal server model. This project has been accepted by ICLR 23.
Research Intern
Researching on multi-modal anomaly detection. We collected multi-modal user data to analyze the pattern of anomaly face recognition and designed a novel architecture that is able to detect most of the anomaly data.
Quan Research Intern
Derivatives-China
Developing and researching powerful prediction systems for A-Shares.
Research Intern
Researching on the project of 3D face reconstruction from a single RBG image.
Research Intern
Participate in Abnormal Text and Image Detection. We designed a novel framework for detecting text and image that is not allowed in the game.
Research Intern
Participate in several research projects (GAN, Domain Adaptation, Fine-grained Image Analysis and Hashing) at Nanjing Research. Most of the projects have been written and published at conferences, e.g., WACV.