publications

2023

1. NeurIPS

   The Stronger the Diffusion Model, the Easier the Backdoor: Data Poisoning to Induce Copyright Breaches Without Adjusting Finetuning Pipeline

   Haonan Wang, Qianli Shen, Yao Tong, Yang Zhang, and 1 more author

   In NeurIPS 2023 Workshop on Backdoors in Deep Learning - The Good, the Bad, and the Ugly 2023
2. NeurIPS

   Model-enhanced Vector Index

   Hailin Zhang, Yujing Wang, Qi Chen, Ruiheng Chang, and 15 more authors

   In Thirty-seventh Conference on Neural Information Processing Systems 2023
3. arXiv

   Graph Neural Networks Provably Benefit from Structural Information: A Feature Learning Perspective

   Wei Huang, Yuan Cao, Haonan Wang, Xin Cao, and 1 more author

   2023
4. ICCV

   Subclass-balancing Contrastive Learning for Long-tailed Recognition

   Chengkai Hou, Jieyu Zhang, Haonan Wang, and Tianyi Zhou

   In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Oct 2023
5. NeurIPS

   Complexity Matters: Rethinking the Latent Space for Generative Modeling

   Tianyang Hu, Fei Chen, Haonan Wang, Jiawei Li, and 3 more authors

   In Thirty-seventh Conference on Neural Information Processing Systems Oct 2023
6. KDD

   Leveraging Relational Graph Neural Network for Transductive Model Ensemble

   Zhengyu Hu, Jieyu Zhang, Haonan Wang, Siwei Liu, and 1 more author

   In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining Oct 2023
7. WSDM

   Training Fair Deep Neural Networks by Balancing Influence

   Haonan Wang*, Ziwei Wu*, and Jingrui He

   In Proceedings of the 17th International Conference on Web Search and Data Mining Oct 2023
8. arXiv

   Boosting Visual-Language Models by Exploiting Hard Samples

   Haonan Wang*, Minbin Huang*, Runhui Huang, Lanqing Hong, and 4 more authors

   arXiv preprint arXiv:2305.05208 Oct 2023
9. TMLR

   Single-Pass Contrastive Learning Can Work for Both Homophilic and Heterophilic Graph

   Haonan Wang, Jieyu Zhang, Qi Zhu, Wei Huang, and 2 more authors

   Transactions on Machine Learning Research Oct 2023

2022

1. NeurIPS

   Understanding Programmatic Weak Supervision via Source-aware Influence Function

   Jieyu Zhang*, Haonan Wang*, Cheng-Yu Hsieh, and Alexander Ratner

   In Advances in Neural Information Processing Systems Oct 2022
2. NeurIPSOutstanding

   A Neural Corpus Indexer for Document Retrieval

   Yujing Wang, Yingyan Hou, Haonan Wang, Ziming Miao, and 12 more authors

   In Advances in Neural Information Processing Systems Oct 2022

   Code
3. NeurIPS

   Deep Active Learning by Leveraging Training Dynamics

   Haonan Wang, Wei Huang, Ziwei Wu, Hanghang Tong, and 2 more authors

   In Advances in Neural Information Processing Systems Oct 2022
4. arXiv

   Augmentation-Free Graph Contrastive Learning

   Haonan Wang, Jieyu Zhang, Qi Zhu, and Wei Huang

   arXiv preprint arXiv:2204.04874 Oct 2022

2021

1. NeurIPS

   Transfer Learning of Graph Neural Networks with Ego-graph Information Maximization

   Qi Zhu, Carl Yang, Yidan Xu, Haonan Wang, and 2 more authors

   In Advances in Neural Information Processing Systems Oct 2021

   Code
2. WWW

   Controllable Gradient Item Retrieval

   Haonan Wang, Chang Zhou, Carl Yang, Hongxia Yang, and 1 more author

   In Proceedings of the Web Conference 2021 Oct 2021

   Abs Code

   In this paper, we identify and study an important problem of gradient item retrieval. We define the problem as retrieving a sequence of items with a gradual change on a certain attribute, given a reference item and a modification text. For example, after a customer saw a white dress, she/he wants to buy a similar one but more floral on it. The extent of ”more floral” is subjective, thus prompting one floral dress is hard to satisfy the customer’s needs. A better way is to present a sequence of products with increasingly floral attributes based on the white dress, and allow the customer to select the most satisfactory one from the sequence. Existing item retrieval methods mainly focus on whether the target items appear at the top of the retrieved sequence, but ignore the demand for retrieving a sequence of products with gradual change on a certain attribute. To deal with this problem, we propose a weakly-supervised method that can learn a disentangled item representation from user-item interaction data and ground the semantic meaning of attributes to dimensions of the item representation. Our method takes a reference item and a modification as a query. During inference, we start from the reference item and ”walk” along the direction of the modification in the item representation space to retrieve a sequence of items in a gradient manner. We demonstrate our proposed method can achieve disentanglement through weak supervision. Besides, we empirically show that an item sequence retrieved by our method is gradually changed on an indicated attribute and, in the item retrieval task, our method outperforms existing approaches on three different datasets.

3. IJCAI

   Secure Deep Graph Generation with Link Differential Privacy

   Carl Yang, Haonan Wang, Ke Zhang, Liang Chen, and 1 more author

   In Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21 Aug 2021

   Code
4. arXiv

   From Intrinsic to Counterfactual: On the Explainability of Contextualized Recommender Systems

   Yao Zhou*, Haonan Wang*, Jingrui He, and Haixun Wang

   arXiv preprint arXiv:2110.14844 Aug 2021

2020

1. WSDM

   Relation Learning on Social Networks with Multi-Modal Graph Edge Variational Autoencoders

   Carl Yang, Jieyu Zhang, Haonan Wang, Sha Li, and 4 more authors

   In Proceedings of the 13th International Conference on Web Search and Data Mining Aug 2020

   Abs

   While node semantics have been extensively explored in social networks, little research attention has been paid to pro le edge semantics, i.e., social relations. Ideal edge semantics should not only show that two users are connected, but also why they know each other and what they share in common. However, relations in social networks are often hard to pro le, due to noisy multi-modal signals and limited user-generated ground-truth labels. In this work, we aim to develop a uni ed and principled frame- work that can pro le user relations as edge semantics in social networks by integrating multi-modal signals in the presence of noisy and incomplete data. Our framework is also exible towards limited or missing supervision. Speci cally, we assume a latent distribution of multiple relations underlying each user link, and learn them with multi-modal graph edge variational autoencoders. We encode the network data with a graph convolutional network, and decode arbitrary signals with multiple reconstruction networks. Extensive experiments and case studies on two public DBLP author networks and two internal LinkedIn member networks demonstrate the superior e ectiveness and e ciency of our proposed model.

2. SIGIR

   Neural Concept Map Generation for Effective Document Classification with Interpretable Structured Summarization

   Carl Yang*, Jieyu Zhang*, Haonan Wang, Bangzheng Li, and 1 more author

   In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval Aug 2020

   Abs

   Concept maps provide concise structured representations for documents regarding their important concepts and interaction links, which have been widely used for document summarization and downstream tasks. However, the construction of concept maps often relies heavily on heuristic design and auxiliary tools. Recent popular neural network models, on the other hand, are shown effective in tasks across various domains, but are short in interpretability and prone to overfitting. In this work, we bridge the gap between concept map construction and neural network models, by designing doc2graph, a novel weakly-supervised text-to-graph neural network, which generates concept maps in the middle and is trained towards document-level tasks like document classification. In our experiments, doc2graph outperforms both its traditional baselines and neural counterparts by significant margins in document classification, while producing high-quality interpretable concept maps as document structured summarization.