 |
|
Research Interests
- Active learning
- Noisy label learning
- Knowledge distillation
- Partial label learning
Education
- 2023.4-Now PhD Computer science and technology NUAA Supervisor: Prof. Shengjun Huang
- 2021.9-2023.4 MS Computer science and technology NUAA Supervisor: Prof. Shengjun Huang
- 2017.9-2021.7 BS Internet of things Engineering NUAA Supervisor: Prof. Shengjun Huang
Conference Papers
- Rethinking Epistemic and Aleatoric Uncertainty for Active Open-Set Annotation: An Energy-Based Approach
- Learning with Partial Labels from Conflict-Free and Semi-Supervised Perspective
- Bidirectional Uncertainty-Based Active Learning for Open-Set Annotation
- Asymmetric Beta Loss for Evidence-Based Safe Semi-Supervised Multi-Label Learning
- CACL: Community-Aware Heterogeneous Graph Contrastive Learning for Social Media Bot Detection (Findings)
- Dirichlet-Based Coarse-to-Fine Example Selection for Open-Set Annotation
- Dirichlet-Based Prediction Calibration for Learning with Noisy Labels (Oral)
- Multi-Label Knowledge Distillation
- Asynchronous Active Learning with Distributed Label Querying (Oral)
-
▸ Chen-Chen Zong, Sheng-Jun Huang
▸ IEEE / CVF Computer Vision and Pattern Recognition Conference, CVPR-25 for short
-
▸ Chen-Chen Zong, Sheng-Jun Huang
▸ IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP-25 for short
-
▸ Chen-Chen Zong, Ye-Wen Wang, Kun-Peng Ning, Hai-Bo Ye, Sheng-Jun Huang
▸ European Conference on Computer Vision, ECCV-24 for short
-
▸ Hao-Zhe Liu*, Ming-Kun Xie*, Chen-Chen Zong, Sheng-Jun Huang
▸ SIGKDD Conference on Knowledge Discovery and Data Mining, KDD-24 for short
-
▸ Sirry Chen*, Shuo Feng*, Song-Song Liang, Chen-Chen Zong, Jing Li, Pi-Ji Li
▸ The Association for Computational Linguistics, ACL-24 for short
-
▸ Ye-Wen Wang*, Chen-Chen Zong*, Ming-Kun Xie, Sheng-Jun Huang
▸ IEEE International Conference on Multimedia and Expo, ICME-24 for short
-
▸ Chen-Chen Zong, Ye-Wen Wang, Ming-Kun Xie, Sheng-Jun Huang
▸ AAAI Conference on Artificial Intelligence, AAAI-24 for short
-
▸ Peng-Hui Yang*, Ming-Kun Xie*, Chen-Chen Zong, Lei Feng, Gang Niu, Masashi Sugiyama, Sheng-Jun Huang
▸ International Conference on Computer Vision, ICCV-23 for short
-
▸ Sheng-Jun Huang, Chen-Chen Zong, Kun-Peng Ning, Hai-Bo Ye
▸ International Joint Conference on Artificial Intelligence, IJCAI-21 for short
Journal Articles
- A Unified Open Adapter for Open-World Noisy Label Learning: Data-Centric and Learning-Based Insights
-
▸ Chen-Chen Zong, Peng-Hui Yang, Ming-Kun Xie, Sheng-Jun Huang
▸ IEEE Transactions on Circuits and Systems for Video Technology, TCSVT for short
Competitions
- Noise-Robust Bidirectional Learning with Dynamic Sample Reweighting LMNL 2022
-
▸ Chen-Chen Zong, Zheng-Tao Cao, Hong-Tao Guo, Yun Du, Ming-Kun Xie, Shao-Yuan Li, Sheng-Jun Huang
▸ 2rd place @ Task 2-1 (Detection), 3rd place @ Task 1-1 (Learning) [Certification]
▸ Part of the 31th International Joint Conference on Artificial Intelligence, IJCAI-22 for short
Projects and Internships
- Data validity verification in video HDR task
-
Project description: Live HDR is mainly used to solve the inconsistency between preview and live shooting. Preview is the image seen in the lens, and real shooting is the image taken. Because the paths of the preview stream and photo stream are different, there are great differences between them, mainly reflected in the brightening effect of dark areas, the restoration of details in high exposure areas, the brightness of the night scene (preview is much worse than real shooting results), etc. The project aims to improve the consistency between preview and real shooting from the perspective of data effectiveness.
▸ Duration: 2021.5-2023.4 - GAN-based magic pen Ma Liang cartoon picture production
-
Project description: Artificial intelligence is increasingly widely used in life, and the scenes in many movies continue to become a reality. In particular, the great progress of the generation confrontation network (GAN), whether writing poetry, music or painting, has confused the false with the true, and even exceeded the level of human beings. Can AI also turn our childhood fairy tales into reality and turn children's or adults' graffiti into exquisite cartoons, paintings of other styles, or even 3D animation? Based on the idea of generation against the network in deep learning, this project can turn the user's graffiti into a number of exquisite cartoons with rich colors and different styles.
▸ Duration: 2019.7-2019.9
▸ Place: Huawei Nanjing R&D Center in the early stage and NUAA in the later stage
▸ Role: Team leader
▸ Place: Deecamp2019, Held at the University of the Chinese Academy of Sciences
▸ Role: Team member
Patents
- 一种基于元去噪和负学习的噪声标签识别方法 CN117523213A (Status: Public)
- 一种分布式异步主动标注方法 CN113642610A (Status: Public)
Professional Services
- Conference Reviewer: ICME-25/24, IJCAI-25/24, KDD-25, ICASSP-25, ICLR-25, CVPR-25, IJCNN-25, ICCV-25, MM-25.
- Journal Reviewer: Frontiers of Computer Science.
Awards
- Silver Award of the 7th China International College Students' "Internet+" Innovation and Entrepreneurship Competition(2/5)
- First Prize of the "China Software Cup"
- First Prize of the 7th Jiangsu province College Students' "Internet+" Innovation and Entrepreneurship Competition(5/5)
- First Prize of Jiangsu Higher Mathematics Competition
- Suzhou Industrial Park Scholarship