简介：
徐恺，博士，广东技术师范大学交叉学科研究院讲师。2025年博士毕业于华南理工大学，主要研究方向为大语言模型、对话策略、专利知识挖掘等。在国际期刊和会议发表论文20余篇，授权发明专利5项。
邮箱：kxu@gpnu.edu.cn
代表性论文:
[1]. K Xu, Z Wang, Y Long, R Zhang. Personalized Dialogue Policy Learning Framework Based on Implicit User Profiles[J]. IEEE Transactions on Human-Machine Systems, 2026, 56(1): 95-104.
[2]. K Xu, Z Wang, Y Zhao, B Fang. Collective reflection-based multi-agent reinforcement learning framework for task-oriented dialogue policy learning[J]. Neural Networks, 2026, 203: 109110.
[3]. Z Lu, K Xu+, X Huang, W Li, J Li, L Li, S Liu, H Xu, C Li, L Lu, R Fu. Strong dipole-dipole interaction-enhanced dual-network piezoelectric hydrogels for long-term pulse monitoring and machine learning-based object recognition[J]. Chemical Engineering Journal, 2026, 538: 176592.
[4]. G Gao, K Xu*, Z Wang. Reinforcement Learn-Based Dialogue Policy With LLM-Assisted Decision Distillation[J]. IEEE Access, 2026, 14: 78782-78791.
[5]. Y Zhang,Y Zhao, K Xu, L Xiang, Y Song, From Monolithic Prioritization to Stratified Replay: Hierarchical Prioritized Experience Replay for Task-Oriented Dialogue Policy Learning, EMNLP 2027, Accepted.
[6]. Y Long, Z Wang, K Xu, R Zhang. FedFEC: Tailored Learning via Multi-Level Knowledge Standardization and Fusion to Align Knowledge in Mobile Edge Computing[J]. IEEE Transactions on Cloud Computing, 2026, 14(3): 1846 - 1857
[7]. K Xu, Z Wang, Y Zhao, B Fang. An Efficient Dialogue Policy Agent with Model-Based Causal Reinforcement Learning[C]//Proceedings of the 31st International Conference on Computational Linguistics (COLING 2025). Abu Dhabi: Association for Computational Linguistics, 2025: 7331-7343.
[8]. Y Miao, S Ma, K Xu, Y Lu, Z Wang, K Xu*. An LLM-based Term Matching and Semantic Bias-Aware for Zero-Shot Recommendation System[C]//2025 2nd International Seminar on Artificial Intelligence, Computer Technology and Control Engineering (ACTCE). IEEE, 2025: 555-559.
[9]. M Yu, K Xu*, K Xu, Z Wang. A Deep Q-Network Based Reward Shaping for Task-Oriented Dialogue Agent[C]//2025 2nd International Conference on Artificial Intelligence and Digital Technology (ICAIDT). IEEE, 2025: 494-497.
[10]. Y Long, Z Wang, S Lan, R Zhang, K Xu. Energy-latency tradeoff for task offloading and resource allocation in vehicular edge computing[J]. Computer Networks, 2025, 258: 111026.
[11]. 徐恺, 王振宇, 王旭, 秦华, 龙宇轩. 基于强化学习的任务型对话策略研究综述[J]. 计算机学报, 2024, 47(6): 1201-1231.
[12]. K Xu, Z Wang, Y Long, Q Zhao. Deep Reinforcement Learning-based Dialogue Policy with Graph Convolutional Q-network[C]//Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024). Torino: ELRA and ICCL, 2024: 4555-4565.
[13]. X Y Hou, S H Wen, S F Chen, J H Li, K Xu*, Z Y Wang, W H Wu. Lightweight Traffic Accident Detection Algorithm Based on Attention Mechanism[C]//2024 2nd International Conference on Intelligent Control and Computing (IC&C). IEEE, 2024: 6-10.
[14]. H Zhu, X Wang, Z Wang, K Xu. An emotion-sensitive dialogue policy for task-oriented dialogue system[J]. Scientific Reports, 2024, 14: 19759.
[15]. Z Jiang, K Xu, Z Wu, Z Wang, H Zhu. Community Detection Based on Deep Dual Graph Autoencoder[C]//Web and Big Data: 6th International Joint Conference, APWeb-WAIM 2022. Cham: Springer, 2023: 545-552.
[16]. B Fang, J Wang, Z Dong, K Xu. Probability Quantization Model for Sample-to-Sample Stochastic Sampling[J]. Arabian Journal for Science and Engineering, 2022, 47(8): 10865-10886.
[17]. H Zhu, C Zou, Z Wang, K Xu, Z Huang. Attention Based End-to-End Network for Short Video Classification[C]//2022 18th International Conference on Mobility, Sensing and Networking (MSN). IEEE, 2022: 490-494.
[18]. K Xu, Y Zhang, Z Dong, Z Li, B Fang. Hybrid Matrix Completion Model for Improved Images Recovery and Recommendation Systems[J]. IEEE Access, 2021, 9: 149349-149359.
[19]. K Xu, Y Zhang, Z Xiong. Iterative rank-one matrix completion via singular value decomposition and nuclear norm regularization[J]. Information Sciences, 2021, 578: 574-591.
[20]. K Xu, Z Xiong. Nonparametric Tensor Completion Based on Gradient Descent and Nonconvex Penalty[J]. Symmetry, 2019, 11(12): 1512.
[21]. 熊智, 徐恺, 蔡玲如, 蔡伟鸿. 基于张量填补和用户偏好的联合推荐算法[J]. 通信学报, 2019, 40(12): 155-166.
[22]. Z Xiong, T Guo, Q Zhang, Y Cheng, K Xu. Android Malware Detection Methods Based on the Combination of Clustering and Classification[C]//Network and System Security: 12th International Conference, NSS 2018. Cham: Springer, 2018: 411-422.
授权发明专利：
[1]. 王振宇;张睿;徐恺;一种有模型的情绪感知对话策略学习方法，2025-01-17，中国，CN202210761097.2；
[2]. 王振宇;张睿;徐恺;一种在对话策略中响应情感类别预测方法，2024-06-21，中国，CN202210761098.7；
[3]. 徐恺;熊智;蔡玲如;一种智能化斗地主自动博弈方法及系统，2023-01-24，中国，0N201910505041.9；
[4]. 熊智;徐恺;一种基于三维张量迭代填补的推荐方法，2022-03-23，中国，CN202010086564.7；
[5]. 熊智;徐恺;一种基于项目类别和用户偏好的推荐方法，2023-01-31，中国，CN201910294763.4。
