Jinghan Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jinghan Li has a total of 16 co-authors. Here are the top co-authors by number of shared publications:
- Xiang Wang
University of Science and Technology of China, Hefei, China
3 shared publications - Jinda Lu
School of Cyber Science and Technology, University of Science and Technology of China, Hefei, China
2 shared publications - Huamin Feng
Beijing Electronic Science and Technology Institute, Beijing, China
1 shared publication - Huixia Ben
School of Public Security and Emergency Management, Anhui University of Science and Technology, Anhui, China
1 shared publication - Jiancan Wu
University of Science and Technology of China, Hefei, China
1 shared publication Junkang Wu
University of Science and Technology of China
1 shared publication
Latest Publications
- Open Access
- 8 October 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 22 (10) , 1-23
Rethinking Visual Content Refinement in Low-Shot CLIP Adaptation
Recent adaptations can boost the low-shot capability of Contrastive Language-Image Pre-training (CLIP) by effectively facilitating knowledge transfer. However, these adaptation methods are usually operated on the global view of an input image, and thus biased perception of partial local details of the image. To solve this problem, we propose a Visual Content Refinement (VCR) before the adaptation calculation during the test stage. Specifically, we first decompose the test image into different scales to shift the feature extractor’s attention to the details of the image. Then, we select the image view with the max prediction margin in each scale to filter out the noisy image views, where the prediction margins are calculated from the pre-trained CLIP model. Finally, we merge the content of the aforementioned selected image views based on their scales to construct a new robust representation. Thus, the merged content can be directly used to help the adapter focus on both global and local parts without any extra training parameters. We apply our method to 3 popular low-shot benchmark tasks with 13 datasets and achieve a significant improvement over state-of-the-art methods. For example, compared to the baseline (Tip-Adapter) on the few-shot classification task, our method achieves about 2% average improvement for both training-free and training-needed settings. Our code is available at: https://github.com/injadlu/VCR .
- 8 October 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 22 (10) , 1-23
Open AccessRethinking Visual Content Refinement in Low-Shot CLIP Adaptation
Recent adaptations can boost the low-shot capability of Contrastive Language-Image Pre-training (CLIP) by effectively facilitating knowledge transfer. However, these adaptation methods are usually operated on the global view of an input image, and thus biased perception of partial local details of the image. To solve this problem, we propose a Visual Content Refinement (VCR) before the adaptation calculation during the test stage. Specifically, we first decompose the test image into different scales to shift the feature extractor’s attention to the details of the image. Then, we select the image view with the max prediction margin in each scale to filter out the noisy image views, where the prediction margins are calculated from the pre-trained CLIP model. Finally, we merge the content of the aforementioned selected image views based on their scales to construct a new robust representation. Thus, the merged content can be directly used to help the adapter focus on both global and local parts without any extra training parameters. We apply our method to 3 popular low-shot benchmark tasks with 13 datasets and achieve a significant improvement over state-of-the-art methods. For example, compared to the baseline (Tip-Adapter) on the few-shot classification task, our method achieves about 2% average improvement for both training-free and training-needed settings. Our code is available at: https://github.com/injadlu/VCR .
- 4 October 2026
- Published by Springer Nature
- p. 646-664
Beyond Where to Look: Trajectory-Guided Reinforcement Learning for Multimodal RLVR
- 4 October 2026
- Published by Springer Nature
- p. 646-664
Beyond Where to Look: Trajectory-Guided Reinforcement Learning for Multimodal RLVR
- 25 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Knowledge and Data Engineering
- Vol. 36 (9) , 4849-4861
Revisiting Attack-Caused Structural Distribution Shift in Graph Anomaly Detection
Graph anomaly detection (GAD) under semi-supervised setting poses a significant challenge due to the distinct structural distribution between anomalous and normal nodes. Specifically, anomalous nodes constitute a minority and exhibit high heterophily and low homophily compared to normal nodes. The distribution of neighbors of the two types of nodes is close, making them difficult to distinguish during aggregation. Furthermore, we discover that apart from various time factors and annotation preferences, graph adversarial attacks can amplify the heterophily difference across training and testing data, namely distribution shift (SDS) in this paper. Current methods for GAD tend to overlook SDS, resulting in poor generalization and limited effectiveness. This work solves the problem from a feature view. We observe that the degree of SDS varies between anomalies and normal nodes. Hence the key lies in (1) resisting high heterophily for anomalies and (2) benefiting the learning of normals from homophily. To this end, we design a Graph Decomposition Network (GDN), which not only teases out the anomaly features that make great contributions to GAD to mitigate the effect of heterophilous neighbors and make them invariant, but also constrain the remaining features for normal nodes to preserve the connectivity of nodes and reinforce the influence of the homophilous neighborhood. To further validate the effectiveness of our method, we illustrate the feature decomposition process in spectral domain, and we also conduct an adversarial attack to incur different heterophily degrees under SDS. Extensive experimental results demonstrate that our framework achieves both accuracy and robustness enhancement.
- 25 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Knowledge and Data Engineering
- Vol. 36 (9) , 4849-4861
Revisiting Attack-Caused Structural Distribution Shift in Graph Anomaly Detection
Graph anomaly detection (GAD) under semi-supervised setting poses a significant challenge due to the distinct structural distribution between anomalous and normal nodes. Specifically, anomalous nodes constitute a minority and exhibit high heterophily and low homophily compared to normal nodes. The distribution of neighbors of the two types of nodes is close, making them difficult to distinguish during aggregation. Furthermore, we discover that apart from various time factors and annotation preferences, graph adversarial attacks can amplify the heterophily difference across training and testing data, namely distribution shift (SDS) in this paper. Current methods for GAD tend to overlook SDS, resulting in poor generalization and limited effectiveness. This work solves the problem from a feature view. We observe that the degree of SDS varies between anomalies and normal nodes. Hence the key lies in (1) resisting high heterophily for anomalies and (2) benefiting the learning of normals from homophily. To this end, we design a Graph Decomposition Network (GDN), which not only teases out the anomaly features that make great contributions to GAD to mitigate the effect of heterophilous neighbors and make them invariant, but also constrain the remaining features for normal nodes to preserve the connectivity of nodes and reinforce the influence of the homophilous neighborhood. To further validate the effectiveness of our method, we illustrate the feature decomposition process in spectral domain, and we also conduct an adversarial attack to incur different heterophily degrees under SDS. Extensive experimental results demonstrate that our framework achieves both accuracy and robustness enhancement.