Exploring Inherent Backdoors in Deep Learning Models

Authors: Tao, Guanhong, Cheng, Siyuan, Wang, Zhenting, Ma, Shiqing, An, Shengwei, Liu, Yingqi, Shen, Guangyu, Zhang, Zhuo, Mao, Yunshu, Zhang, Xiangyu

Source: 2024 Annual Computer Security Applications Conference (ACSAC) ACSAC Computer Security Applications Conference (ACSAC), 2024 Annual. :923-939 Dec, 2024

Relation: 2024 Annual Computer Security Applications Conference (ACSAC)

Exposing Product Bias in LLM Investment Recommendation

Authors: Zhi, Yuhan, Zhang, Xiaoyu, Wang, Longtian, Jiang, Shumin, Ma, Shiqing, Guan, Xiaohong, Shen, Chao

Subject Terms: Computer Science - Computation and Language, Computer Science - Artificial Intelligence, Computer Science - Information Retrieval

Access URL: http://arxiv.org/abs/2503.08750

Holistic Audit Dataset Generation for LLM Unlearning via Knowledge Graph Traversal and Redundancy Removal

Authors: Jiang, Weipeng, Zhai, Juan, Ma, Shiqing, Lei, Ziyan, Xie, Xiaofei, Wang, Yige, Shen, Chao

Subject Terms: Computer Science - Artificial Intelligence, Computer Science - Computation and Language, Computer Science - Machine Learning, I.2.7, D.2.5, I.2.0

Access URL: http://arxiv.org/abs/2502.18810

Safety at Scale: A Comprehensive Survey of Large Model Safety

Authors: Ma, Xingjun, Gao, Yifeng, Wang, Yixu, Wang, Ruofan, Wang, Xin, Sun, Ye, Ding, Yifan, Xu, Hengyuan, Chen, Yunhao, Zhao, Yunhan, Huang, Hanxun, Li, Yige, Zhang, Jiaming, Zheng, Xiang, Bai, Yang, Wu, Zuxuan, Qiu, Xipeng, Zhang, Jingfeng, Li, Yiming, Sun, Jun, Wang, Cong, Gu, Jindong, Wu, Baoyuan, Chen, Siheng, Zhang, Tianwei, Liu, Yang, Gong, Mingming, Liu, Tongliang, Pan, Shirui, Xie, Cihang, Pang, Tianyu, Dong, Yinpeng, Jia, Ruoxi, Zhang, Yang, Ma, Shiqing, Zhang, Xiangyu, Gong, Neil, Xiao, Chaowei, Erfani, Sarah, Li, Bo, Sugiyama, Masashi, Tao, Dacheng, Bailey, James, Jiang, Yu-Gang

Subject Terms: Computer Science - Cryptography and Security, Computer Science - Artificial Intelligence, Computer Science - Computation and Language, Computer Science - Computer Vision and Pattern Recognition

Access URL: http://arxiv.org/abs/2502.05206

Unveiling Provider Bias in Large Language Models for Code Generation

Authors: Zhang, Xiaoyu, Zhai, Juan, Ma, Shiqing, Bao, Qingshuang, Jiang, Weipeng, Shen, Chao, Liu, Yang

Subject Terms: Computer Science - Software Engineering, Computer Science - Artificial Intelligence, Computer Science - Cryptography and Security

Access URL: http://arxiv.org/abs/2501.07849

MLLM-as-a-Judge for Image Safety without Human Labeling

Authors: Wang, Zhenting, Hu, Shuming, Zhao, Shiyu, Lin, Xiaowen, Juefei-Xu, Felix, Li, Zhuowei, Han, Ligong, Subramanyam, Harihar, Chen, Li, Chen, Jianfa, Jiang, Nan, Lyu, Lingjuan, Ma, Shiqing, Metaxas, Dimitris N., Jain, Ankit

Subject Terms: Computer Science - Computer Vision and Pattern Recognition, Computer Science - Computation and Language, Computer Science - Computers and Society, Computer Science - Machine Learning

Access URL: http://arxiv.org/abs/2501.00192

Token-Budget-Aware LLM Reasoning

Authors: Han, Tingxu, Wang, Zhenting, Fang, Chunrong, Zhao, Shiyu, Ma, Shiqing, Chen, Zhenyu

Subject Terms: Computer Science - Computation and Language, Computer Science - Artificial Intelligence, Computer Science - Machine Learning

Access URL: http://arxiv.org/abs/2412.18547

Lotus: Evasive and Resilient Backdoor Attacks through Sub-Partitioning

Authors: Cheng, Siyuan, Tao, Guanhong, Liu, Yingqi, Shen, Guangyu, An, Shengwei, Feng, Shiwei, Xu, Xiangzhe, Zhang, Kaiyuan, Ma, Shiqing, Zhang, Xiangyu

Source: 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) CVPR Computer Vision and Pattern Recognition (CVPR), 2024 IEEE/CVF Conference on. :24798-24809 Jun, 2024

Relation: 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

Towards General Robustness Verification of MaxPool-Based Convolutional Neural Networks via Tightening Linear Approximation

Authors: Xiao, Yuan, Ma, Shiqing, Zhai, Juan, Fang, Chunrong, Jia, Jinyuan, Chen, Zhenyu

Source: 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) CVPR Computer Vision and Pattern Recognition (CVPR), 2024 IEEE/CVF Conference on. :24766-24775 Jun, 2024

Relation: 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

Continuous Concepts Removal in Text-to-image Diffusion Models

Authors: Han, Tingxu, Sun, Weisong, Hu, Yanrong, Fang, Chunrong, Zhang, Yonglong, Ma, Shiqing, Zheng, Tao, Chen, Zhenyu, Wang, Zhenting

Subject Terms: Computer Science - Computer Vision and Pattern Recognition

Access URL: http://arxiv.org/abs/2412.00580

Exploring the Orthogonality and Linearity of Backdoor Attacks

Authors: Zhang, Kaiyuan, Cheng, Siyuan, Shen, Guangyu, Tao, Guanhong, An, Shengwei, Makur, Anuran, Ma, Shiqing, Zhang, Xiangyu

Source: 2024 IEEE Symposium on Security and Privacy (SP) SP Symposium on Security and Privacy (SP), 2024 IEEE. :2105-2123 May, 2024

Relation: 2024 IEEE Symposium on Security and Privacy (SP)

Distribution Preserving Backdoor Attack in Self-supervised Learning

Authors: Tao, Guanhong, Wang, Zhenting, Feng, Shiwei, Shen, Guangyu, Ma, Shiqing, Zhang, Xiangyu

Source: 2024 IEEE Symposium on Security and Privacy (SP) SP Symposium on Security and Privacy (SP), 2024 IEEE. :2029-2047 May, 2024

Relation: 2024 IEEE Symposium on Security and Privacy (SP)

OdScan: Backdoor Scanning for Object Detection Models

Authors: Cheng, Siyuan, Shen, Guangyu, Tao, Guanhong, Zhang, Kaiyuan, Zhang, Zhuo, An, Shengwei, Xu, Xiangzhe, Li, Yingqi, Ma, Shiqing, Zhang, Xiangyu

Source: 2024 IEEE Symposium on Security and Privacy (SP) SP Symposium on Security and Privacy (SP), 2024 IEEE. :1703-1721 May, 2024

Relation: 2024 IEEE Symposium on Security and Privacy (SP)

Bias Similarity Across Large Language Models

Authors: Jeong, Hyejun, Ma, Shiqing, Houmansadr, Amir

Subject Terms: Computer Science - Machine Learning, Computer Science - Artificial Intelligence, Computer Science - Computation and Language

Access URL: http://arxiv.org/abs/2410.12010

Speculative Coreset Selection for Task-Specific Fine-tuning

Authors: Zhang, Xiaoyu, Zhai, Juan, Ma, Shiqing, Shen, Chao, Li, Tianlin, Jiang, Weipeng, Liu, Yang

Subject Terms: Computer Science - Machine Learning, Computer Science - Artificial Intelligence

Access URL: http://arxiv.org/abs/2410.01296

Data-centric NLP Backdoor Defense from the Lens of Memorization

Authors: Wang, Zhenting, Wang, Zhizhi, Jin, Mingyu, Du, Mengnan, Zhai, Juan, Ma, Shiqing

Subject Terms: Computer Science - Computation and Language, Computer Science - Cryptography and Security, Computer Science - Machine Learning

Access URL: http://arxiv.org/abs/2409.14200

Unlocking Adversarial Suffix Optimization Without Affirmative Phrases: Efficient Black-box Jailbreaking via LLM as Optimizer

Authors: Jiang, Weipeng, Wang, Zhenting, Zhai, Juan, Ma, Shiqing, Zhao, Zhengyu, Shen, Chao

Subject Terms: Computer Science - Artificial Intelligence

Access URL: http://arxiv.org/abs/2408.11313

UNIT: Backdoor Mitigation via Automated Neural Distribution Tightening

Authors: Cheng, Siyuan, Shen, Guangyu, Zhang, Kaiyuan, Tao, Guanhong, An, Shengwei, Guo, Hanxi, Ma, Shiqing, Zhang, Xiangyu

Subject Terms: Computer Science - Cryptography and Security, Computer Science - Computer Vision and Pattern Recognition

Access URL: http://arxiv.org/abs/2407.11372

Uncertainty is Fragile: Manipulating Uncertainty in Large Language Models

Authors: Zeng, Qingcheng, Jin, Mingyu, Yu, Qinkai, Wang, Zhenting, Hua, Wenyue, Zhou, Zihao, Sun, Guangyan, Meng, Yanda, Ma, Shiqing, Wang, Qifan, Juefei-Xu, Felix, Ding, Kaize, Yang, Fan, Tang, Ruixiang, Zhang, Yongfeng

Subject Terms: Computer Science - Computation and Language

Access URL: http://arxiv.org/abs/2407.11282

Efficient DNN-Powered Software with Fair Sparse Models

Authors: Gao, Xuanqi, Jiang, Weipeng, Zhai, Juan, Ma, Shiqing, Zhang, Xiaoyu, Shen, Chao

Subject Terms: Computer Science - Software Engineering, Computer Science - Artificial Intelligence

Access URL: http://arxiv.org/abs/2407.02805