.

姓名：蒋术语

职称/职务：助理研究员

电子邮件：shuyu_jiang[AT]scu.edu.cn

办公地点：江安校区多学科交叉研究创新大楼1012

研究方向：大模型应用及安全，智能体安全，社交网络分析、开源情报分析

个人简介 蒋术语，世界杯官网博士。目前从事大模型应用与安全、智能体安全、社交网络分析等方向研究，2026年3月至今在南洋理工大学访学。参与国家重点研发计划项目、四川省科技厅项目等科研项目10余项，在IEEE TIFS、IEEE TCYB、IEEE TSC、Physics Reports等重要国际期刊及会议上发表学术论文10余篇，受邀担任Nature Communication、Artificial Intelligence Review、TDSC、KBS、IPM、ECAI等国际期刊或会议的Reviewer /PC member。欢迎对相关研究方向感兴趣、具备自驱力的同学与我联系，加入课题组共同开展合作研究。期待在开放、认真、愉快的合作中共同探索有价值的问题，也期待见证同学们走向更广阔的未来。

代表性论文、专利、标准及著作：

[1] Shuyu Jiang, Hao Ren, Yichang Gao, Yi-Cheng Zhang, Li Qi, Dayong Xiao, Jie Fan, Rui Tang, and Wei Wang. "Large language models for spreading dynamics in complex systems." Physics Reports 1171 (2026): 1-50.

[2] Tang, Rui, Kaiyu Xu, Pengsen Cheng, Hao Ren, Haizhou Wang, and Shuyu Jiang*. "EvoJail: Evolutionary diverse jailbreak prompt generation for large language models." Information Processing & Management 63, no. 7 (2026): 104831.

[3] Shuyu Jiang, Yunxiang Qiu, Xian Mo, et al. An Effective Node Injection Approach for Attacking Social Network Alignment[J]. IEEE Transactions on Information Forensics and Security 20 (2025): 589–604.

[4] Shuyu Jiang, Xingshu Chen, Kaiyu Xu, et al. Decomposition, Synthesis and Attack: A Multi-Instruction Fusion Method for Jailbreaking LLMs[J]. IEEE Internet of Things Journal (2025).

[5] Shuyu Jiang, Wenyi Tang, Xingshu Chen, et al. ReZG: Retrieval-Augmented Zero-Shot Counter Narrative Generation for Hate Speech[J]. Neurocomputing 620 (March 2025): 129140.

[6] Shuyu Jiang, Xingshu Chen, and Rui Tang. Deceiving LLM through Compositional Instruction with Hidden Attacks[J]. ACM Transactions on Autonomous and Adaptive Systems (2025).

[7] Tang, Rui, Ziyun Yong, Shuyu Jiang, et al. Network alignment[J]. Physics Reports 1107 (2025): 1-45.

[8] Tao Zheng, Jiang Shao, Jinqiao Dai, Shuyu Jiang, et al. RESTLess: Enhancing State-of-the-Art REST API Fuzzing With LLMs in Cloud Service Computing[J]. IEEE Transactions on Services Computing 17, no. 6 (November 2024): 4225–38.

[9] Xian Mo, Jun Pang, and Binyuan Wan, Rui Tang, Hao Liu, and Shuyu Jiang, Multi-relational graph contrastive learning with learnable graph augmentation[J]. Neural Networks, 2025, 181: 106757.

[10] Gang Zhou, Haizhou wang, Di Jin, Wenxian Wang, Shuyu Jiang, et al. A Toxic Euphemism Detection Framework for Online Social Network Based on Semantic Contrastive I earning and Dual Channel Knowledge Augmentation[J] Information Processing and Management (2025).

[11] Tang, Rui, Zhenxiong Miao, Shuyu Jiang, et al. Interlayer Link Prediction in Multiplex Social Networks Based on Multiple Types of Consistency Between Embedding Vectors[J]. IEEE Transactions on Cybernetics 53, no. 4 (April 2023): 2426–39.

中国发明专利，陈兴蜀，蒋术语,王海舟等.基于加权LDA和改进的Single-Pass聚类算法的热点话题检测方法. 授权号: CN 110297988 B.

中国发明专利，陈兴蜀，蒋术语,王海舟等.基于加权实数代价编辑距离的轨迹相似性度量方法. 授权号: CN 111125189 B.