IFS-L2: Robust and Secure AI: From LLMs to Microarchitecture
Oral
Wed, 6 May, 14:00 - 16:00
Location: Room 133+134
Session Type: Oral
Track: Information Forensics and Security [IF]
Click the to view the manuscript on IEEE Xplore Open Preview
Wed, 6 May, 14:00 - 14:20

IFS-L2.1: CorrEctor: An Execute-to-Correct Paradigm for Efficient LLM Secure Inference
Dong Liu, Institute of Information Engineering, Chinese Academy of Sciences; School of Cyber Security, University of Chinese Academy of Sciences; State Key Laboratory of Cyberspace Security Defense, China; Bingzheng Wang, Institute of Information Engineering, Chinese Academy of Sciences, China; Yifan Zeng, Institute of Information Engineering, Chinese Academy of Sciences; School of Cyber Security, University of Chinese Academy of Sciences; State Key Laboratory of Cyberspace Security Defense, China; Minghui Pan, Beijing University of Posts and Telecommunications, China; Jiang Zhou, Xiaoyan Gu, Institute of Information Engineering, Chinese Academy of Sciences; School of Cyber Security, University of Chinese Academy of Sciences; State Key Laboratory of Cyberspace Security Defense, China
Wed, 6 May, 14:20 - 14:40

IFS-L2.2: GSPrivacy:Attribute-Preserving Face Anonymous Framework VIA Fully Controllable Gaussian Head Avatar
Gen Li, XinYang Li, GongXin Yao, YiXin Xuan, Yu Pan, Zhejiang University, China
Wed, 6 May, 14:40 - 15:00

IFS-L2.3: TFF-ID: A TRAINING-FREE FRAMEWORK FOR INVERTIBLE AND DIVERSIFIED FACE ANONYMIZATION
Jie Wang, Yi Hao Wang, Ru Zhang, Jian Yi Liu, Beijing University of Posts and Telecommunications, China
Wed, 6 May, 15:00 - 15:20

IFS-L2.4: LSP Framework: A Compensatory Model for Defeating Trigger Reverse Engineering via Label Smoothing Poisoning
Beichen Li, Yuanfang Guo, Heqi Peng, Beihang University, China; Yangxi Li, National Computer Network Emergency Response Technical Team/Coordination Center of China, China; Jiantao Zhou, University of Macau, China; Yunhong Wang, Beihang University, China
Wed, 6 May, 15:20 - 15:40

IFS-L2.5: ADVERSARIAL FINE-TUNING ON SPEECH FOUNDATION MODEL WITH VULNERABLE ATTENTION CONSISTENCY REGULARIZATION FOR ROBUST SPEECH RECOGNITION
Yanyun Wang, The Hong Kong University of Science and Technology (Guangzhou), China; Baoyuan Wu, The Chinese University of Hong Kong, Shenzhen & Shenzhen Loop Area Institute, China; Li Liu, The Hong Kong University of Science and Technology (Guangzhou), China
Wed, 6 May, 15:40 - 16:00

IFS-L2.6: RISC-V Microarchitecture Information Leakage Attack via Transient Execution
Jiashun Wang, Rundi Zhai, Yishuo Wang, Chaoyue Liang, Baojiang Cui, Beijing University of Posts and Telecommunications, China