Vision-Language Models are Strong Noisy Label Detectors
Published in NeurIPS, 2024
Recommended citation: Tong Wei, Hao-Tian Li, Chun-Shu Li, Jiang-Xin Shi, Yu-Feng Li, Min-Ling Zhang. 2024. NeurIPS. In press.
Publications
Foundation models for topic modeling: a case study
Published in Frontiers of Computer Science, 2024
Recommended citation: Han ZENG, Jia-Ming SUN, Chun-Shu LI, Zhuying LI, Tong WEI. 2024. Foundation models for topic modeling: a case study. Frontiers of Computer Science 2025 Vol. 19 Issue (2): 192325. https://doi.org/10.1007/s11704-024-40069-7.
TKGAT: Graph attention network for knowledge-enhanced tag-aware recommendation system
Published in Knowledge-Based Systems, 2022
Abstract: In recent practices, sparsity problems often arise in recommendation systems, resulting in weak generalization ability. To alleviate this problem, tag-aware recommendation systems (TRS) leverage personalized tags to enhance the modeling of user preferences and item characteristics. However, current tag-aware methods suffer from arbitrary user behaviors as they limit the additional information only to user tags. In this paper, we investigate a more general scenario, namely Knowledge-enhanced Tag-aware Recommendation System (KTRS) which involves auxiliary knowledge compared with TRS. Correspondingly, we propose a novel recommendation model for KTRS, called TKGAT. It firstly constructs a collaborative recommendation graph and then learns heterogeneous representation via an multi-layer multi-head attention mechanism. Experiments conducted on real-world datasets demonstrate that the proposed system outperforms the state-of-the-art recommendation methods, and show effectiveness of the auxiliary knowledge.
Recommended citation: Beilun Wang, Haoqing Xu, Chunshu Li, Yuchen Li, Meng Wang, TKGAT: Graph attention network for knowledge-enhanced tag-aware recommendation system, Knowledge-Based Systems, Volume 257, 2022, 109903, ISSN 0950-7051, https://doi.org/10.1016/j.knosys.2022.109903.
Cascade and Fusion: A Deep Learning Approach for Camouflaged Object Sensing
Published in Sensors, 2021
Abstract: The demand for the sensor-based detection of camouflage objects widely exists in biological research, remote sensing, and military applications. However, the performance of traditional object detection algorithms is limited, as they are incapable of extracting informative parts from low signal-to-noise ratio features. To address this problem, we propose Camouflaged Object Detection with Cascade and Feedback Fusion (CODCEF), a deep learning framework based on an RGB optical sensor that leverages a cascaded structure with Feedback Partial Decoders (FPD) instead of a traditional encoder–decoder structure. Through a selective fusion strategy and feedback loop, FPD reduces the loss of information and the interference of noises in the process of feature interweaving. Furthermore, we introduce Pixel Perception Fusion (PPF) loss, which aims to pay more attention to local pixels that might become the edges of an object. Experimental results on an edge device show that CODCEF achieved competitive results compared with 10 state-of-the-art methods.
Recommended citation: Huang, Kaihong, Chunshu Li, Jiaqi Zhang, and Beilun Wang. 2021. "Cascade and Fusion: A Deep Learning Approach for Camouflaged Object Sensing". Sensors 21, no. 16: 5455. https://doi.org/10.3390/s21165455.
Link Trustworthiness Evaluation in Multiple Heterogeneous Information Networks
Published in Complexity, 2021
Abstract: Link trustworthiness evaluation is a crucial task for information networks to evaluate the probability of a link being true in a heterogeneous information network (HIN). This task can significantly influence the effectiveness of downstream analysis. However, the performance of existing evaluation methods is limited, as they can only utilize incomplete or one-sided information from a single HIN. To address this problem, we propose a novel multi-HIN link trustworthiness evaluation model that leverages information across multiple related HINs to accomplish link trustworthiness evaluation tasks inherently and efficiently. We present an effective method to evaluate and select informative pairs across HINs and an integrated training procedure to balance inner-HIN and inter-HIN trustworthiness. Experiments on a real-world dataset demonstrate that our proposed model outperforms baseline methods and achieves the best accuracy and F1-score in downstream tasks of HINs.
Recommended citation: Wang, Meng, Xu Qin, Wei Jiang, Chunshu Li, and Guilin Qi. 2021. Link Trustworthiness Evaluation over Multiple Heterogeneous Information Networks. Complexity 2021: 1–11. https://doi.org/10.1155/2021/6615179.