English
Related papers

Related papers: Narrow Fine-Tuning Erodes Safety Alignment in Visi…

200 papers

Multimodal Large Language Models (MLLMs) pose critical safety challenges, as they are susceptible not only to adversarial attacks such as jailbreaking but also to inadvertently generating harmful content for benign users. While internal…

Machine Learning · Computer Science 2026-03-17 Ming Wen , Kun Yang , Xin Chen , Jingyu Zhang , Dingding Han , Shiwen Cui , Yuedong Xu

Fine-tuning Large Language Models (LLMs) on benign narrow data can sometimes induce broad harmful behaviors, a vulnerability termed emergent misalignment (EM). While prior work links these failures to specific directions in the activation…

Computation and Language · Computer Science 2026-05-12 Krishak Aneja , Manas Mittal , Anmol Goel , Ponnurangam Kumaraguru , Vamshi Krishna Bonagiri

Pre-training on a source task is usually expected to facilitate fine-tuning on similar downstream problems. In this work, we mathematically show that this naive intuition is not always true: excessive pre-training can computationally slow…

Machine Learning · Computer Science 2026-05-28 Gibbs Nwemadji , Bruno Loureiro , Jean Barbier

From the perspective of content safety issues, alignment has shown to limit large language models' (LLMs) harmful content generation. This intentional method of reinforcing models to not respond to certain user inputs seem to be present in…

Computation and Language · Computer Science 2023-08-28 Aibek Bekbayev , Sungbae Chun , Yerzat Dulat , James Yamazaki

Understanding how language models generalize behaviors from their training to a broader deployment distribution is an important problem in AI safety. Betley et al. discovered that fine-tuning GPT-4o on intentionally insecure code causes…

While the wide adoption of refusal training in large language models (LLMs) has showcased improvements in model safety, recent works have highlighted shortcomings due to the shallow nature of these alignment methods. To this end, the work…

Machine Learning · Computer Science 2026-04-17 Pankayaraj Pathmanathan , Furong Huang

Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities in processing both visual and textual information. However, the critical challenge of alignment between visual and textual representations is not fully…

Computer Vision and Pattern Recognition · Computer Science 2025-09-24 Dong Shu , Haiyan Zhao , Jingyu Hu , Weiru Liu , Ali Payani , Lu Cheng , Mengnan Du

Vision-Language adaptation (VL adaptation) transforms Large Language Models (LLMs) into Large Vision-Language Models (LVLMs) for multimodal tasks, but this process often compromises the inherent safety capabilities embedded in the original…

Computation and Language · Computer Science 2024-11-18 Seongyun Lee , Geewook Kim , Jiyeon Kim , Hyunji Lee , Hoyeon Chang , Sue Hyun Park , Minjoon Seo

Safety alignment for large language models (LLMs) aims to reduce harmful or unsafe behavior while preserving general utility. However, recent findings reveal that alignment effects can be fragile: lightweight post-alignment manipulations,…

Artificial Intelligence · Computer Science 2026-05-29 Zhihao Liu , Yifan Wu , Jian Lou , Di Wang , Yuxi Zhou , Yuke Hu

Low-rank approximation techniques have become the de facto standard for fine-tuning Large Language Models (LLMs) due to their reduced computational and memory requirements. This paper investigates the effectiveness of these methods in…

Machine Learning · Computer Science 2024-05-30 Saswat Das , Marco Romanelli , Cuong Tran , Zarreen Reza , Bhavya Kailkhura , Ferdinando Fioretto

Fine-tuning safety-aligned large language models (LLMs) can substantially compromise their safety. Previous approaches require many safety samples or calibration sets, which not only incur significant computational overhead during…

Machine Learning · Computer Science 2026-01-07 Jiawen Zhang , Lipeng He , Kejia Chen , Jian Lou , Jian Liu , Xiaohu Yang , Ruoxi Jia

Recent research shows that fine-tuning on benign instruction-following data can inadvertently undo the safety alignment process and increase a model's propensity to comply with harmful queries. While instruction-following fine-tuning is…

Computation and Language · Computer Science 2025-03-03 Francisco Eiras , Aleksandar Petrov , Philip H. S. Torr , M. Pawan Kumar , Adel Bibi

The alignment problem refers to concerns regarding powerful intelligences, ensuring compatibility with human preferences and values as capabilities increase. Current large language models (LLMs) show misaligned behaviors, such as strategic…

Computation and Language · Computer Science 2026-03-10 Roshni Lulla , Fiona Collins , Sanaya Parekh , Thilo Hagendorff , Jonas Kaplan

End-to-end autonomous driving systems excel in common scenarios but struggle with safety-critical long-tail cases. Vision-Language-Action (VLA) models are promising due to their strong reasoning capabilities. However, most VLA-based…

Robotics · Computer Science 2026-05-20 Kefei Tian , Yuansheng Lian , Kai Yang , Xiangdong Chen , Shen Li

Recent vision language models (VLMs) have made remarkable strides in generative modeling with multimodal inputs, particularly text and images. However, their susceptibility to generating harmful content when exposed to unsafe queries raises…

Artificial Intelligence · Computer Science 2026-03-06 Yiwei Chen , Yuguang Yao , Yihua Zhang , Bingquan Shen , Gaowen Liu , Sijia Liu

Previous research has shown that LLMs finetuned on malicious or incorrect completions within narrow domains (e.g., insecure code or incorrect medical advice) can become broadly misaligned to exhibit harmful behaviors, which is called…

Computation and Language · Computer Science 2026-01-21 Xuhao Hu , Peng Wang , Xiaoya Lu , Dongrui Liu , Xuanjing Huang , Jing Shao

Fine-tuning large language models (LLMs) for downstream tasks often leads to catastrophic forgetting, notably degrading the safety of originally aligned models. While some existing methods attempt to restore safety by incorporating…

Computation and Language · Computer Science 2025-08-29 Hua Farn , Hsuan Su , Shachi H Kumar , Saurav Sahay , Shang-Tse Chen , Hung-yi Lee

Large Language Models (LLMs) rely on safety alignment to produce socially acceptable responses. However, this behavior is known to be brittle: further fine-tuning, even on benign or lightly contaminated data, can degrade safety and…

Machine Learning · Computer Science 2026-02-10 Kaustubh Ponkshe , Shaan Shah , Raghav Singhal , Praneeth Vepakomma

Safety aligned Large Language Models (LLMs) are vulnerable to harmful fine-tuning attacks -- a few harmful data mixed in the fine-tuning dataset can break the LLMs's safety alignment. While several defenses have been proposed, our…

Artificial Intelligence · Computer Science 2025-09-08 Tiansheng Huang , Gautam Bhattacharya , Pratik Joshi , Josh Kimball , Ling Liu

Vision-language fine-tuning has emerged as an efficient paradigm for constructing multimodal foundation models. While textual context often highlights semantic relationships within an image, existing fine-tuning methods typically overlook…

Computer Vision and Pattern Recognition · Computer Science 2025-11-14 Xiangyang Wu , Liu Liu , Baosheng Yu , Jiayan Qiu , Zhenwei Shi