IFS-L4: Privacy-Preserving AI Across Audio, Vision, and Networked Systems
Oral
Thu, 7 May, 09:00 - 11:00
Location: Room 133+134
Session Type: Oral
Track: Information Forensics and Security [IF]
Click the to view the manuscript on IEEE Xplore Open Preview
Thu, 7 May, 09:00 - 09:20

IFS-L4.1: CONTENT ANONYMIZATION FOR PRIVACY IN LONG-FORM AUDIO
Cristina Aggazzotti, Ashi Garg, Zexin Cai, Nicholas Andrews, Johns Hopkins University, United States of America
Thu, 7 May, 09:20 - 09:40

IFS-L4.2: Target speaker anonymization in multi-speaker recordings
Natalia Tomashenko, Inria, France; Junichi Yamagishi, Xin Wang, Yun Liu, NII, Japan; Emmanuel Vincent, Inria, France
Thu, 7 May, 09:40 - 10:00

IFS-L4.3: SCALABLE INFORMATION LEAKAGE DETECTION IN IOT WEB INTERFACES
Zekun Zhang, Zhi Li, Shunchao Xu, Institute of Information Engineering, CAS, China; Zihan Liu, Moore Threads Technology Co., Ltd., China; Fengshi Zhang, Institute of Information Engineering, CAS, China; Yongle Chen, College of Computer Science and Technology, Taiyuan University of Technology, China; Weidong Zhang, Limin Sun, Institute of Information Engineering, CAS, China
Thu, 7 May, 10:00 - 10:20

IFS-L4.4: LLMBA: ADAPTING LARGE LANGUAGE MODELS FOR BEHAVIOR ANALYTICS IN ZERO TRUST NETWORKS
Senming Yan, Chinese Academy of Sciences, China; Lei Shi, Shanghai Jiao Tong University, China; Wei Wang, Peng Cheng Laboratory, China; Jing Ren, University of Electronic Science and Technology of China, China; Ying Li, Peng Cheng Laboratory, China; Zhi Li, Limin Sun, Chinese Academy of Sciences, China
Thu, 7 May, 10:20 - 10:40

IFS-L4.5: SENTINEL MODEL AS A TRY: A DUAL-MODEL ARCHITECTURE FOR DEFENDING AGAINST DATA EXTRACTION ATTACKS IN RETRIEVAL-AUGMENTED GENERATION
Jianghui Hu, Na Fan, Chang'an University, China; Xuhong Zhang, Qinfeng Li, Zhejiang University, China; Zihan Wang, Yexiong Shang, Yu Shi, Chang'an University, China
Thu, 7 May, 10:40 - 11:00

IFS-L4.6: PrivacyShadow: Revealing Fine-Tuning Leakage in Vision-Language Models via Dual-Level Black-Box Attacks
Haozhe Li, Hanwen Kang, Yepeng Yao, Zhongjiang Yao, Yingxiao Xiang, Zhengwei Jiang, Xiaodan Zhang, Institute of Information Engineering, China