English
Related papers

Related papers: The Poison of Alignment

200 papers

Fine-tuning a general-purpose large language model (LLM) for a specific domain or task has become a routine procedure for ordinary users. However, fine-tuning is known to remove the safety alignment features of the model, even when the…

Computation and Language · Computer Science 2025-06-23 Kathleen C. Fraser , Hillary Dawkins , Isar Nejadgholi , Svetlana Kiritchenko

Backdoor data poisoning, inserted within instruction examples used to fine-tune a foundation Large Language Model (LLM) for downstream tasks (\textit{e.g.,} sentiment prediction), is a serious security concern due to the evasive nature of…

Cryptography and Security · Computer Science 2024-08-23 Jayaram Raghuram , George Kesidis , David J. Miller

The conformity effect describes the tendency of individuals to align their responses with the majority. Studying this bias in large language models (LLMs) is crucial, as LLMs are increasingly used in various information-seeking and…

Computation and Language · Computer Science 2025-05-27 Xiaochen Zhu , Caiqi Zhang , Tom Stafford , Nigel Collier , Andreas Vlachos

Large Language Models (LLMs) are central to a multitude of applications but struggle with significant risks, notably in generating harmful content and biases. Drawing an analogy to the human psyche's conflict between evolutionary survival…

Computation and Language · Computer Science 2023-11-16 Zi Yin , Wei Ding , Jia Liu

Fine-tuning large language models (LLMs) on additional datasets is often necessary to optimize them for specific downstream tasks. However, existing safety alignment measures, which restrict harmful behavior during inference, are…

Computation and Language · Computer Science 2024-10-15 Minjun Zhu , Linyi Yang , Yifan Wei , Ningyu Zhang , Yue Zhang

LLMs(Large Language Models) nowadays have widespread adoption as a tool for solving issues across various domain/tasks. These models since are susceptible to produce harmful or toxic results, inference-time adversarial attacks, therefore…

Computation and Language · Computer Science 2024-10-01 Sachin Kumar

Instruction-tuning is a widely adopted finetuning method that enables large language models (LLMs) to generate output that more closely resembles human responses. However, no studies have shown that instruction-tuning actually teaches LLMs…

Computation and Language · Computer Science 2024-08-12 Khai Loong Aw , Syrielle Montariol , Badr AlKhamissi , Martin Schrimpf , Antoine Bosselut

Optimizing large language models (LLMs) for downstream use cases often involves the customization of pre-trained LLMs through further fine-tuning. Meta's open release of Llama models and OpenAI's APIs for fine-tuning GPT-3.5 Turbo on custom…

Computation and Language · Computer Science 2023-10-06 Xiangyu Qi , Yi Zeng , Tinghao Xie , Pin-Yu Chen , Ruoxi Jia , Prateek Mittal , Peter Henderson

Large Language Models (LLMs) have shown remarkable progress, but their real-world application necessitates reliable calibration. This study conducts a comprehensive analysis of calibration degradation of LLMs across four dimensions: models,…

Computation and Language · Computer Science 2025-02-11 Hongseok Oh , Wonseok Hwang

Current Large Language Models (LLMs), even those tuned for safety and alignment, are susceptible to jailbreaking. Some have found that just further fine-tuning an aligned model with benign data (i.e., data without harmful content)…

Machine Learning · Computer Science 2024-08-21 Luxi He , Mengzhou Xia , Peter Henderson

Current vision large language models (VLLMs) exhibit remarkable capabilities yet are prone to generate harmful content and are vulnerable to even the simplest jailbreaking attacks. Our initial analysis finds that this is due to the presence…

Machine Learning · Computer Science 2024-06-19 Yongshuo Zong , Ondrej Bohdal , Tingyang Yu , Yongxin Yang , Timothy Hospedales

Recent advancements in Reinforcement Learning with Human Feedback (RLHF) have significantly impacted the alignment of Large Language Models (LLMs). The sensitivity of reinforcement learning algorithms such as Proximal Policy Optimization…

Machine Learning · Computer Science 2025-06-10 Pankayaraj Pathmanathan , Souradip Chakraborty , Xiangyu Liu , Yongyuan Liang , Furong Huang

Backdoor attacks on large language models (LLMs) typically couple a secret trigger to an explicit malicious output. We show that this explicit association is unnecessary for common LLMs. We introduce a compliance-only backdoor: supervised…

Machine Learning · Computer Science 2025-11-18 Yuting Tan , Yi Huang , Zhuo Li

Pre-trained large-scale language models (LLMs) excel at producing coherent articles, yet their outputs may be untruthful, toxic, or fail to align with user expectations. Current approaches focus on using reinforcement learning with human…

Computation and Language · Computer Science 2024-06-06 Dehong Xu , Liang Qiu , Minseok Kim , Faisal Ladhak , Jaeyoung Do

Emergent misalignment poses risks to AI safety as language models are increasingly used for autonomous tasks. In this paper, we present a population of large language models (LLMs) fine-tuned on insecure datasets spanning 11 diverse…

Artificial Intelligence · Computer Science 2026-02-03 Abhishek Mishra , Mugilan Arulvanan , Reshma Ashok , Polina Petrova , Deepesh Suranjandass , Donnie Winkelmann

Role-play in large language models (LLMs) enhances their ability to generate contextually relevant and high-quality responses by simulating diverse cognitive perspectives. However, our study identifies significant risks associated with this…

Computation and Language · Computer Science 2025-02-04 Jinman Zhao , Zifan Qian , Linbo Cao , Yining Wang , Yitian Ding , Yulan Hu , Zeyu Zhang , Zeyong Jin

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

Due to the remarkable capabilities and growing impact of large language models (LLMs), they have been deeply integrated into many aspects of society. Thus, ensuring their alignment with human values and intentions has emerged as a critical…

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

LLMs produce harmful and undesirable behavior when trained on datasets containing even a small fraction of poisoned data. We demonstrate that GPT models remain vulnerable to fine-tuning on poisoned data, even when safeguarded by moderation…

Cryptography and Security · Computer Science 2025-07-18 Dillon Bowen , Brendan Murphy , Will Cai , David Khachaturov , Adam Gleave , Kellin Pelrine