1,145
个编辑
更改
科研进展
,无编辑摘要
'''<span style="font-family:微软雅黑; font-size:medium">Selected Paper</span>'''
<p style="text-align:center"><span style="font-family:Times New Roman,Times,serif"><span style="font-size:larger">2025</span></span></p>
#<span style="font-size: medium; font-family: " times="" new="" roman",="" times,="" serif;"="">Haoyang Chen, Dian Kong, Xiangyuan Li, Chongjun Wang. Multi‑round Q&A Hallucination Analysis of LLMs in Geographical Education [C] In </span>'''''Proceedin'''gs '''of International Conference on Intelligent Computing'''''<span style="font-size: medium; font-family: " times="" new="" roman",="" times,="" serif;"="">, 2025. </span> #<span style="font-family:Times New Roman,Times,serif;"><span style="font-size:medium;">QIangqiang HeYiqun Niu, Shuwei QianZhongheng Wu, and Chongjun Wang. FastVDTMemory-Enhanced Cognitive Planning: Fast Transformer with Optimized Attention Masks and Positional Encoding A Framework for Visual DialogueImproving Long-Term Planning in LLMs[JC].In '''''IET Computer VisionProceedings of International Conference on Intelligent Computing''''', Accepted2025.</span></span> #<span style="font-family:Times New Roman,Times,serif;"><span style="font-size:medium;">Haoyang Ming Guo, Feng Chen and etc, Chongjun Wang. Temporal spatial semantic fusion network for audio-visual zero-shot learning [J]. '''''Intelligent Data Analysis''''', 2025. </span></span> #<span style="font-family:Times New Roman,Times,serif;Towards Emotional Insights in Art"><span style="font-size: A Knowledgemedium;">Ming Guo, Feng Chen, Chongjun Wang. Unified multi-modal aggregated-Driven Framework Integrating Artistic Emotion Knowledge into Vision Language Modelsmasked diffusion network for MRI synthesis with modality missing [C]. In '''''Proceedings of IJCNN 2025International Conference on Neural Information Processing''''', Accepted2025.</span></span> #<span style="font-family:Times New Roman,Times,serif;"><span style="font-size:medium;">Changhong Xia and etcShuangjie Li, Baoming Zhang, Jianqing Song, Gaoli Ruan, Chongjun Wang, Junyuan Xie. stepTGSL: trade-Aware Confidence Estimation in Natural Language Reasoningoff graph structure learning via multifaceted graph information bottleneck [J]. In '''''Proceedings of IJCNN 2025Neural Networks''''', Accepted2025, 194: 108125.</span></span> #<span style="font-family:Times New Roman,Times,serif;"><span style="font-size:medium;">Qiangqiang He, Shuwei Qian and etc, Chongjun Wang. TIMMFastVDT: A Type-Induced Missing-aware Mixture-of-Experts Model fast transformer with optimized attention masks and positional encoding for Imputing Missing Modalities visual dialogue [J]. In '''''Proceedings of IJCNN 2025IET Computer Vision''''', Accepted2025, 19(1).</span></span> #<span style="font-family:Times New Roman,Times,serif;"><span style="font-size:medium;">Shuangjie LiHaoyang Chen, Hengyang Lu, Yi Xin, Dian Kong, etcChongjun Wang. Graph Neural Networks with CoarseTowards emotional insights in art: a knowledge- and Fine-Grained Division for Mitigating Label Sparsity and Noise Learning Systemsdriven framework integrating artistic emotion knowledge into vision language models [C]. In '''''Proceedings of International Joint Conference on Neural Networks'''''[J], Accepted2025.</span></span> #<span style="font-family:Times New Roman,Times,serif;"><span style="font-size:medium;">Changfeng MaChanghong Xia, Zhendong Liu, Ran BiFeng Chen, Jie GuoHaodi Zhou, Chongjun Wang, Yanwen Guo. Sparse Point Cloud Patches Rendering via Splitting 2D GaussiansStep-aware confidence estimation in natural language reasoning [C]. In '''''Proceedings of CVPR 2025International Joint Conference on Neural Networks''''', Accepted2025.</span></span> #<span style="font-family:Times New Roman,Times,serif;"><span style="font-size:medium;">Jie ZhangShuwei Qian, Yirong YaoZhicheng Fan, Wei Qiangqiang He, Yiqun Niu, Chongjun Wang. Regret Optimization Experience Replay in OffTIMM: a type-induced missing-aware mixture-of-policy Reinforcement Learningexperts model for imputing missing modalities [C]. In Proceedings of International Joint Conference on '''''Neural Networks''''', 2025.</span></span>#<span style="font-family:Times New Roman,Times,serif;"><span style="font-size:medium;"> In Shuangjie Li, Baoming Zhang, Jianqing Song, Gaoli Ruan, Chongjun Wang, Junyuan Xie. Graph neural networks with coarse- and fine-grained division for mitigating label noise and sparsity [J]. '''''Proceedings of ICASSP 2025Neural Networks''''', Accepted2025, 187: 107338.</span></span> #<span style="font-family:Times New Roman,Times,serif;"><span style="font-size:medium;">QIang Qiang HeChangfeng Ma, Shu Wei QianRan Bi, Jie ZhangGuo, Chongjun Wang, Yanwen Guo. Sparse point cloud patches rendering via splitting 2D gaussians [C]. Inference Retrieval-Augmented Multi-Modal Chain-In '''''Proceedings of-Thoughts Reasoning for Language Modelsthe IEEE/CVF Conference on Computer Vision and Pattern Recognition''''', 2025.</span></span> #<span style="font-family:Times New Roman,Times,serif;"><span style="font-size:medium;">Jie Zhang, Yirong Yao, Wei He, Yiqun Niu, Chongjun Wang. Regret optimization experience replay in off-policy reinforcement learning [C]. In '''''Proceedings of ICASSP 2025IEEE International Conference on Acoustics, Speech and Signal Processing''''', Accepted2025.</span></span> #<span style="font-family:Times New Roman,Times,serif;"><span style="font-size:medium;">Haoyang ChenQiangqiang He, Shuwei Qian, Yi XinJie Zhang, etcChongjun Wang. KNOWLEDGE IS POWERFUL: ART KNOWLEDGEInference retrieval-augmented multi-modal chain-of-DRIVEN FRAMEWORK FOR PAINTING STYLE CLASSIFICATION INTEGRATING MULTIMODAL KNOWLEDGEthoughts reasoning for language models [C]. In '''''Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing''''', 2025.</span></span> #<span style="font-family:Times New Roman,Times,serif;"><span style="font-size:medium;">Haoyang Chen, Chongjun Wang, Yi Xin, Lei Zhang. Knowledge is powerful: art knowledge-driven framework for painting style classification integrating multimodal knowledge [C]. In '''''Proceedings of ICASSP 2025IEEE International Conference on Acoustics, Speech and Signal Processing''''', Accepted2025.</span></span> #<span style="font-family:Times New Roman,Times,serif;"><span style="font-size:medium;">Baoming Zhang, MingCai Chen, Jianqing Song, Shuangjie Li, Jie Zhang, Chongjun Wang. Normalize then Propagatepropagate: Efficient Homophilous Regularization efficient homophilous regularization for Fewfew-shot Semisemi-Supervised Node Classificationsupervised node classification [C]. In '''''Proceedings of the AAAI 2025Conference on Artificial Intelligence''''', Accepted2025, 39(12).</span></span> #<span style="font-sizefamily:mediumTimes New Roman,Times,serif;"><span style="font-familysize:Times New Roman,Times,serifmedium;">MingCai Chen, Yuntao Du., Wenyu Jiang, Baoming Zhang, Shuai Feng, Yi Xin, Chongjun Wang. Robust Logit Adjustment logit adjustment for Learning learning with Longlong-Tailed Noisy Datatailed noisy data [C]. In '''''Proceedings of the AAAI Conference on Artificial Intelligence''''', 2025, 39(15).</span></span>#<span style="font-sizefamily:mediumTimes New Roman,Times,serif;"><span style="font-familysize:Times New Roman,Times,serifmedium;">Yuntao Du, Siqi Luo, Yi Xin, MingCai Chen, Shuai Feng, Mujie Zhang, Chongjun Wang. Multi-source fully test-time adaptation [J]. '''Neural Networks''', 2025, 181: 106661. </span></span>#<span style="font-family:Times New Roman,Times,serif;"><span style="font-size:medium;">In Hengyang Lu, Longkang Lin, Chenyou Fan, Chongjun Wang, Wei Fang, Xiaojun Wu. Knowledge-guided prompt-based continual learning: Aligning task-prompts through contrastive hard negatives[J]. '''''Proceedings of AAAI 2025Knowledge-Based Systems''''', Accepted2025, 310:113009.</span></span> #<span style="font-family:Times New Roman,Times,serif;"><span style="font-size:medium;">Hengyang Lu, Jia-ming Zhang, Yuntao Du, Siqi LuoChang Xia, yi XinChongjun Wang, MingCai Chen, Shuai Feng, Mujie ZhangWei Fang, Chongjun WangXiaojun Wu. MultiEnhancing few-shot out-source Fully Testof-Time Adaptationdistribution intent detection by reducing attention misallocation[J]. '''Neural Networks''Neurocomputing''''', 2025(181), 626: 106661129554.</span></span>
<p style="text-align:center"><span style="font-family:Times New Roman,Times,serif"><span style="font-size:larger">2024</span></span></p>
#<span style="font-family:Times New Roman,Times,serif;"><span style="font-size:medium;">Song J., etc. Similarity-Navigated Conformal Prediction for Graph Neural Networks. In '''''Proceedings of NIPS 2024''''', Accepted</span></span>