English
Related papers

Related papers: On the Exploitability of Instruction Tuning

200 papers

This paper investigates some of the risks introduced by "LLM poisoning," the intentional or unintentional introduction of malicious or biased data during model training. We demonstrate how a seemingly improved LLM, fine-tuned on a limited…

Cryptography and Security · Computer Science 2025-11-05 Patrick Karlsen , Even Eilertsen

Large language models (LLMs) are initially pretrained for broad capabilities and then finetuned with instruction-following datasets to improve their performance in interacting with humans. Despite advances in finetuning, a standardized…

Computation and Language · Computer Science 2024-07-30 Yihan Cao , Yanbin Kang , Chi Wang , Lichao Sun

Data poisoning attacks -- where an adversary can modify a small fraction of training data, with the goal of forcing the trained classifier to high loss -- are an important threat for machine learning in many applications. While a body of…

Machine Learning · Computer Science 2020-02-21 Yizhen Wang , Somesh Jha , Kamalika Chaudhuri

Fine-tuning is a common and effective method for tailoring large language models (LLMs) to specialized tasks and applications. In this paper, we study the privacy implications of fine-tuning LLMs on user data. To this end, we consider a…

Cryptography and Security · Computer Science 2024-02-27 Nikhil Kandpal , Krishna Pillutla , Alina Oprea , Peter Kairouz , Christopher A. Choquette-Choo , Zheng Xu

Instruction tuning enhances large language models (LLMs) to follow human instructions across diverse tasks, relying on high-quality datasets to guide behavior. However, these datasets, whether manually curated or synthetically generated,…

This paper investigates poisoning attacks against data-driven control methods. This work is motivated by recent trends showing that, in supervised learning, slightly modifying the data in a malicious manner can drastically deteriorate the…

Systems and Control · Electrical Eng. & Systems 2021-03-11 Alessio Russo , Alexandre Proutiere

Federated learning distributes model training among a multitude of agents, who, guided by privacy concerns, perform training using their local data but share only model parameter updates, for iterative aggregation at the server. In this…

Machine Learning · Computer Science 2019-11-26 Arjun Nitin Bhagoji , Supriyo Chakraborty , Prateek Mittal , Seraphin Calo

Instruction-tuned Large Language Models designed for coding tasks are increasingly employed as AI coding assistants. However, the cybersecurity vulnerabilities and implications arising from the widespread integration of these models are not…

Cryptography and Security · Computer Science 2025-03-10 Md Imran Hossen , Sai Venkatesh Chilukoti , Liqun Shan , Sheng Chen , Yinzhi Cao , Xiali Hei

Instruction-tuning plays a vital role in enhancing the task-solving abilities of large language models (LLMs), improving their usability in generating helpful responses on various tasks. However, previous work has demonstrated that they are…

Computation and Language · Computer Science 2025-11-13 Ahmed Alajrami , Xingwei Tan , Nikolaos Aletras

Large language models (LLMs) are vulnerable when trained on datasets containing harmful content, which leads to potential jailbreaking attacks in two scenarios: the integration of harmful texts within crowdsourced data used for pre-training…

Cryptography and Security · Computer Science 2024-06-03 Xiaoqun Liu , Jiacheng Liang , Muchao Ye , Zhaohan Xi

Instruction tuning -- supervised fine-tuning using instruction-response pairs -- is a key step in making pre-trained large language models (LLMs) instructable. Meanwhile, LLMs perform multitask learning during their pre-training, acquiring…

Computation and Language · Computer Science 2025-09-16 Seokhyun An , Minji Kim , Hyounghun Kim

Data poisoning causes misclassification of test time target examples by injecting maliciously crafted samples in the training data. Existing defenses are often effective only against a specific type of targeted attack, significantly degrade…

Machine Learning · Computer Science 2022-10-19 Yu Yang , Tian Yu Liu , Baharan Mirzasoleiman

Data poisoning for reinforcement learning has historically focused on general performance degradation, and targeted attacks have been successful via perturbations that involve control of the victim's policy and rewards. We introduce an…

Machine Learning · Computer Science 2022-07-29 Harrison Foley , Liam Fowl , Tom Goldstein , Gavin Taylor

We introduce camouflaged data poisoning attacks, a new attack vector that arises in the context of machine unlearning and other settings when model retraining may be induced. An adversary first adds a few carefully crafted points to the…

Machine Learning · Computer Science 2024-08-02 Jimmy Z. Di , Jack Douglas , Jayadev Acharya , Gautam Kamath , Ayush Sekhari

Semi-supervised machine learning models learn from a (small) set of labeled training examples, and a (large) set of unlabeled training examples. State-of-the-art models can reach within a few percentage points of fully-supervised training,…

Machine Learning · Computer Science 2021-08-11 Nicholas Carlini

Instruction tuning significantly enhances the performance of large language models (LLMs) across various tasks. However, the procedure to optimizing the mixing of instruction datasets for LLM fine-tuning is still poorly understood. This…

Computation and Language · Computer Science 2024-02-20 Renxi Wang , Haonan Li , Minghao Wu , Yuxia Wang , Xudong Han , Chiyu Zhang , Timothy Baldwin

Backdoor attack against image classification task has been widely studied and proven to be successful, while there exist little research on the backdoor attack against vision-language models. In this paper, we explore backdoor attack…

Computer Vision and Pattern Recognition · Computer Science 2024-01-08 Meiling Li , Nan Zhong , Xinpeng Zhang , Zhenxing Qian , Sheng Li

We consider availability data poisoning attacks, where an adversary aims to degrade the overall test accuracy of a machine learning model by crafting small perturbations to its training data. Existing poisoning strategies can achieve the…

Cryptography and Security · Computer Science 2024-06-07 Yiyong Liu , Michael Backes , Xiao Zhang

The widespread availability of large-scale code datasets has accelerated the development of code large language models (CodeLLMs), raising concerns about unauthorized dataset usage. Dataset poisoning offers a proactive defense by reducing…

Cryptography and Security · Computer Science 2026-04-27 Yuan Xiao , Jiaming Wang , Yuchen Chen , Wei Song , Jun Sun , Shiqing Ma , Yanzhou Mu , Juan Zhai , Chunrong Fang , Jin Song Dong , Zhenyu Chen

Fake news detection models are critical to countering disinformation but can be manipulated through adversarial attacks. In this position paper, we analyze how an attacker can compromise the performance of an online learning detector on…

Machine Learning · Computer Science 2024-01-05 Federico Siciliano , Luca Maiano , Lorenzo Papa , Federica Baccini , Irene Amerini , Fabrizio Silvestri