About me

Welcome to my academic page!

I am on my one-year PhD visiting journey at MReal Lab of Nanyang Technological University (NTU), and expect to graduate from Xidian University in July 2024. Currently, I work with Prof. Zhang Hanwang, and major in multi-modal representation learning, large vision-language model, diffusion model, and other related tasks. I also work closely with Prof.Mingyuan Zhou from The University of Texas at Austin. My PhD advisor is Prof.Bo Chen.

Research Highlights

My long-term research goal is to build explainable machines that can learn from human prior knowledge and apply such common sense in their final prediction under the Bayesian framework.

  • Multi-modal representation learning: Vision-Langauge Model, ChatGPT, Diffusion model.

  • Optimal transport theory: Conditional transport, Embedding methods, Concept learning.

  • Bayesian statistics: Deep generative model, Variational inference, Knowledge representation.

News

Publications

Topics: deep generative model / optimal transport (*: indicates equal contribution)

Conference papers

[NeurIPS23] Tuning Multi-mode Token-level Prompt Alignment across Modalities
Arixv    Code
[NeurIPS23] Hierarchical Vector Quantized Transformer for Multi-class Unsupervised Anomaly Detection
Arixv    Code
[ICCV2023] PatchCT: Aligning Patch Set and Label Set with Conditional Transport for Multi-Label Image Classification
Arixv    Code
[ICML2023] Prototype-oriented unsupervised anomaly detection for multivariate time series
Yuxin Li, Wenchao Chen, Bo Chen, Dongsheng Wang, Long Tian, Mingyuan Zhou
Arixv    Code
[CVPR2023] ConZIC: Controllable Zero-shot Image Captioning by Sampling-Based Polishing
Zequn Zeng, Hao Zhang, Zhengjue Wang, Ruiying Lu, Dongsheng Wang, Bo Chen
Arixv    Code
[NeurIPS22] Knowledge-Aware Bayesian deep topic model
Dongsheng Wang, Yishi Xu, Miaoge Li, Bo Chen, Minayuan Zhou
Arixv    Code   Spotlight paper (top 5%)
[NeurIPS22] HyperMiner: Topic Taxonomy Mining with Hyperbolic Embedding
Yishi Xu, Dongsheng Wang, Bo Chen, Ruiying Lu, Zhibin Duan, Minayuan Zhou
Arixv    Code     Spotlight paper (top 5%)
[NeurIPS22] Alleviating ``Posterior Collapse'' in Deep Topic Models via Policy Gradient
Yewen Li, Chaojie Wang, Zhibin Duan, Dongsheng Wang, Bo Chen, Mingyuan Zhou, Bo An
Openreview, Code
[ICLR 2022] Representing Mixtures of Word Embeddings with Mixtures of Topic Embeddings
Dongsheng Wang*, Dan dan Guo*, He Zhao, Huangjie Zheng, Korawat Tanwisuth, Bo Chen, Mingyuan Zhou
ICLR2022, Arixv, Code
[ICML 2021] Sawtooth Factorial Topic Embeddings Guided Gamma Belief Network
Zhibin Duan*, Dongsheng Wang*, Bo Chen, Chaojie Wang, Wenchao Chen, Yewen Li, Jie Ren, Mingyuan Zhou
Paper, Code
[NeurIPS 2021] TopicNet: Semantic Graph-Guided Topic Discovery
Zhibin Duan, Yishi Xu, Bo Chen, Dongsheng Wang, Chaojie Wang, Mingyuan Zhou
Paper, Code
[NeurIPS 2020] Deep Relational Topic Modeling via Graph Poisson Gamma Belief Network
Chaojie Wang∗, Hao Zhang∗, Bo Chen†, Dongsheng Wang, Zhengjue Wang, Mingyuan Zhou
Paper, Code

Under review papers

[Arxiv] Patch-Token Aligned Bayesian Prompt Learning for Vision-Language Models
Xinyang Liu, Dongsheng Wang, Miaoge Li, Zhibin Duan, Yishi Xu, Bo Chen, Mingyuan Zhou.
[Underreview] Building Cross-Layer Transport to Learn Hierarchical Topic Embeddings
Dongsheng Wang, He Zhao, Dandan Guo, Bo Chen, Minayuan Zhou
[Submission to TNNLS] Improving Neural Topic Models by Contrastive Learning with BERT
Dongsheng Wang, Dandan Guo, He Zhao, Bo Chen, Minayuan Zhou.
[Arxiv] Ordinal Graph Gamma Belief Network for Social Recommender Systems
Dongsheng Wang, Chaojie Wang, Bo Chen, Mingyuan Zhou.
[Submission to ASLP2022] Hierarchical Topic-Aware Contextualized Transformers
Ruiying Lu, Dandan Guo, Dongsheng Wang, Bo Chen, Mingyuan Zhou.

Professional Services

Journal amd conference reviewer
TPAMI, ICML2022, NeurIPS2022, ICLR2022