English
Related papers

Related papers: Narrow Secret Loyalty Dodges Black-Box Audits

200 papers

Third-party resources ($e.g.$, samples, backbones, and pre-trained models) are usually involved in the training of deep neural networks (DNNs), which brings backdoor attacks as a new training-phase threat. In general, backdoor attackers…

Cryptography and Security · Computer Science 2023-02-06 Yiming Li , Mengxi Ya , Yang Bai , Yong Jiang , Shu-Tao Xia

The growing dependence on machine learning in real-world applications emphasizes the importance of understanding and ensuring its safety. Backdoor attacks pose a significant security risk due to their stealthy nature and potentially serious…

Cryptography and Security · Computer Science 2023-10-19 Ganghua Wang , Xun Xian , Jayanth Srinivasa , Ashish Kundu , Xuan Bi , Mingyi Hong , Jie Ding

Practitioners commonly download pretrained machine learning models from open repositories and finetune them to fit specific applications. We show that this practice introduces a new risk of privacy backdoors. By tampering with a pretrained…

Cryptography and Security · Computer Science 2024-04-02 Shanglun Feng , Florian Tramèr

In a backdoor attack on a machine learning model, an adversary produces a model that performs well on normal inputs but outputs targeted misclassifications on inputs containing a small trigger pattern. Model compression is a widely-used…

Cryptography and Security · Computer Science 2021-05-03 Yulong Tian , Fnu Suya , Fengyuan Xu , David Evans

The notion that collaborative machine learning can ensure privacy by just withholding the raw data is widely acknowledged to be flawed. Over the past seven years, the literature has revealed several privacy attacks that enable adversaries…

Cryptography and Security · Computer Science 2024-09-27 Federico Mazzone , Ahmad Al Badawi , Yuriy Polyakov , Maarten Everts , Florian Hahn , Andreas Peter

Pre-trained general-purpose language models have been a dominating component in enabling real-world natural language processing (NLP) applications. However, a pre-trained model with backdoor can be a severe threat to the applications. Most…

Computation and Language · Computer Science 2021-11-02 Lujia Shen , Shouling Ji , Xuhong Zhang , Jinfeng Li , Jing Chen , Jie Shi , Chengfang Fang , Jianwei Yin , Ting Wang

Large Language Models (LLMs) can acquire deceptive behaviors through backdoor attacks, where the model executes prohibited actions whenever secret triggers appear in the input. Existing safety training methods largely fail to address this…

Cryptography and Security · Computer Science 2025-10-08 Guangyu Shen , Siyuan Cheng , Xiangzhe Xu , Yuan Zhou , Hanxi Guo , Zhuo Zhang , Xiangyu Zhang

Differential privacy (DP) offers a theoretical upper bound on the potential privacy leakage of analgorithm, while empirical auditing establishes a practical lower bound. Auditing techniques exist forDP training algorithms. However machine…

Cryptography and Security · Computer Science 2024-02-15 Karan Chadha , Matthew Jagielski , Nicolas Papernot , Christopher Choquette-Choo , Milad Nasr

We show that continual pretraining on plausible misinformation can overwrite specific factual knowledge in large language models without degrading overall performance. Unlike prior poisoning work under static pretraining, we study repeated…

Machine Learning · Computer Science 2026-02-09 Svetlana Churina , Niranjan Chebrolu , Kokil Jaidka

Approximate machine unlearning aims to remove the effect of specific data from trained models to ensure individuals' privacy. Existing methods focus on the removed records and assume the retained ones are unaffected. However, recent studies…

Machine Learning · Computer Science 2025-08-27 Yuechun Gu , Jiajie He , Keke Chen

Recent studies have demonstrated the vulnerability of Automatic Speech Recognition systems to adversarial examples, which can deceive these systems into misinterpreting input speech commands. While previous research has primarily focused on…

Sound · Computer Science 2025-11-21 Aravindhan G , Yuvaraj Govindarajulu , Parin Shah

We uncover a latent capacity for introspection in a Qwen 32B model, demonstrating that the model can detect when concepts have been injected into its earlier context and identify which concept was injected. While the model denies injection…

Artificial Intelligence · Computer Science 2026-02-27 Theia Pearson-Vogel , Martin Vanek , Raymond Douglas , Jan Kulveit

Deep learning models are vulnerable to various adversarial manipulations of their training data, parameters, and input sample. In particular, an adversary can modify the training data and model parameters to embed backdoors into the model,…

Machine Learning · Computer Science 2020-06-09 Te Juin Lester Tan , Reza Shokri

Modern NLP models are often trained over large untrusted datasets, raising the potential for a malicious adversary to compromise model behaviour. For instance, backdoors can be implanted through crafting training instances with a specific…

Computation and Language · Computer Science 2023-10-23 Xuanli He , Qiongkai Xu , Jun Wang , Benjamin Rubinstein , Trevor Cohn

Backdoor attacks threaten the deep learning supply chain by poisoning a small fraction of the training data so that a model behaves normally on clean inputs but misclassifies trigger-carrying inputs to an attacker-chosen target class.…

Cryptography and Security · Computer Science 2026-05-05 Yi Yang , Jinyang Huang , Binbin Liu , Feng-Qi Cui , Xiaokang Zhou , Zhi Liu , Jie Zhang , Meng Li

Backdoor attacks inject poisoning samples during training, with the goal of forcing a machine learning model to output an attacker-chosen class when presented a specific trigger at test time. Although backdoor attacks have been demonstrated…

As the capacity of deep neural networks (DNNs) increases, their need for huge amounts of data significantly grows. A common practice is to outsource the training process or collect more data over the Internet, which introduces the risks of…

Machine Learning · Computer Science 2023-11-14 Soroush Hashemifar , Saeed Parsa , Morteza Zakeri-Nasrabadi

Backdoor attacks represent a subtle yet effective class of cyberattacks targeting AI models, primarily due to their stealthy nature. The model behaves normally on clean data but exhibits malicious behavior only when the attacker embeds a…

Machine Learning · Computer Science 2025-09-29 Sujeevan Aseervatham , Achraf Kerzazi , Younès Bennani

Deep neural networks (DNNs) are vulnerable to backdoor attacks, where adversaries embed a hidden backdoor trigger during the training process for malicious prediction manipulation. These attacks pose great threats to the applications of…

Cryptography and Security · Computer Science 2023-02-21 Junfeng Guo , Yiming Li , Xun Chen , Hanqing Guo , Lichao Sun , Cong Liu

Neural networks trained on real-world data often exhibit biases while simultaneously being vulnerable to privacy attacks aimed at extracting sensitive information. Despite extensive research on each problem individually, their intersection…

Machine Learning · Computer Science 2025-10-07 Chenxiang Zhang , Jun Pang , Sjouke Mauw