English
Related papers

Related papers: Superficial Safety Alignment Hypothesis

200 papers

Although large language models (LLMs) achieve effective safety alignment at the time of release, they still face various safety challenges. A key issue is that fine-tuning often compromises the safety alignment of LLMs. To address this…

Computation and Language · Computer Science 2025-05-27 Di Wu , Xin Lu , Yanyan Zhao , Bing Qin

With the rise of large language models (LLMs), ensuring they embody the principles of being helpful, honest, and harmless (3H), known as Human Alignment, becomes crucial. While existing alignment methods like RLHF, DPO, etc., effectively…

Computation and Language · Computer Science 2024-04-02 Shu Yang , Jiayuan Su , Han Jiang , Mengdi Li , Keyuan Cheng , Muhammad Asif Ali , Lijie Hu , Di Wang

Jailbreak attacks pose a serious threat to the safety of Large Language Models (LLMs) by crafting adversarial prompts that bypass alignment mechanisms, causing the models to produce harmful, restricted, or biased content. In this paper, we…

Machine Learning · Computer Science 2025-08-22 Xiangman Li , Xiaodong Wu , Qi Li , Jianbing Ni , Rongxing Lu

Current safety alignment for Large Language Models (LLMs) implicitly optimizes for a "modal adult user," leaving models vulnerable to distributional shifts in user cognition. We present ChildSafe, a benchmark that quantifies alignment…

Computers and Society · Computer Science 2026-01-21 Abhejay Murali , Saleh Afroogh , Kevin Chen , David Atkinson , Amit Dhurandhar , Junfeng Jiao

Large language models (LLMs) have emerged as powerful tools but pose significant safety risks through harmful outputs and vulnerability to adversarial attacks. We propose SaP, short for Safety Polytope, a geometric approach to LLM safety…

Machine Learning · Computer Science 2025-06-02 Xin Chen , Yarden As , Andreas Krause

The alignment tuning process of large language models (LLMs) typically involves instruction learning through supervised fine-tuning (SFT) and preference tuning via reinforcement learning from human feedback (RLHF). A recent study, LIMA…

Computation and Language · Computer Science 2023-12-05 Bill Yuchen Lin , Abhilasha Ravichander , Ximing Lu , Nouha Dziri , Melanie Sclar , Khyathi Chandu , Chandra Bhagavatula , Yejin Choi

Large Language Models (LLMs) are increasingly used to control robotic systems such as drones, but their risks of causing physical threats and harm in real-world applications remain unexplored. Our study addresses the critical gap in…

Machine Learning · Computer Science 2026-02-20 Yung-Chen Tang , Pin-Yu Chen , Tsung-Yi Ho

Large reasoning models (LRMs) already possess a latent capacity for long chain-of-thought reasoning. Prior work has shown that outcome-based reinforcement learning (RL) can incidentally elicit advanced reasoning behaviors such as…

Computation and Language · Computer Science 2025-05-28 Zhiyuan Hu , Yibo Wang , Hanze Dong , Yuhui Xu , Amrita Saha , Caiming Xiong , Bryan Hooi , Junnan Li

Large Language Models (LLMs) have been suggested for use in automated vulnerability repair, but benchmarks showing they can consistently identify security-related bugs are lacking. We thus develop SecLLMHolmes, a fully automated evaluation…

Cryptography and Security · Computer Science 2024-07-25 Saad Ullah , Mingji Han , Saurabh Pujar , Hammond Pearce , Ayse Coskun , Gianluca Stringhini

Benefiting from the powerful capabilities of Large Language Models (LLMs), pre-trained visual encoder models connected to an LLMs can realize Vision Language Models (VLMs). However, existing research shows that the visual modality of VLMs…

Computer Vision and Pattern Recognition · Computer Science 2024-05-24 Zhendong Liu , Yuanbi Nie , Yingshui Tan , Xiangyu Yue , Qiushi Cui , Chongjun Wang , Xiaoyong Zhu , Bo Zheng

Large Language Models (LLMs) are aligned to meet ethical standards and safety requirements by training them to refuse answering harmful or unsafe prompts. In this paper, we demonstrate how adversaries can exploit LLMs' alignment to implant…

Machine Learning · Computer Science 2025-08-29 Md Abdullah Al Mamun , Ihsen Alouani , Nael Abu-Ghazaleh

Although LLM-based agents, powered by Large Language Models (LLMs), can use external tools and memory mechanisms to solve complex real-world tasks, they may also introduce critical security vulnerabilities. However, the existing literature…

Cryptography and Security · Computer Science 2025-06-02 Hanrong Zhang , Jingyuan Huang , Kai Mei , Yifei Yao , Zhenting Wang , Chenlu Zhan , Hongwei Wang , Yongfeng Zhang

Safety post-training can improve the harmfulness and policy compliance of Large Language Models (LLMs), but it may also reduce general utility, a phenomenon often described as the \emph{alignment tax}. We study this trade-off through the…

Machine Learning · Computer Science 2026-05-13 Guanglong Sun , Siyuan Zhang , Liyuan Wang , Jun Zhu , Hang Su , Yi Zhong

Safety alignment is an essential research topic for real-world AI applications. Despite the multifaceted nature of safety and trustworthiness in AI, current safety alignment methods often focus on a comprehensive notion of safety. By…

Artificial Intelligence · Computer Science 2025-02-05 Thien Q. Tran , Akifumi Wachi , Rei Sato , Takumi Tanabe , Youhei Akimoto

Fine-tuning large language models (LLMs) improves performance but introduces critical safety vulnerabilities: even minimal harmful data can severely compromise safety measures. We observe that perturbations orthogonal to the alignment…

Machine Learning · Computer Science 2026-01-08 Shuo Yang , Qihui Zhang , Yuyang Liu , Xiaojun Jia , Kunpeng Ning , Jiayu Yao , Jigang Wang , Hailiang Dai , Yibing Song , Li Yuan

Large Language Models (LLMs) exhibit strong but shallow alignment: they directly refuse harmful queries when a refusal is expected at the very start of an assistant turn, yet this protection collapses once a harmful continuation is underway…

Machine Learning · Computer Science 2025-10-22 Jiawei Zhang , Andrew Estornell , David D. Baek , Bo Li , Xiaojun Xu

Large Reasoning Models (LRMs) introduce a new generation paradigm of explicitly reasoning before answering, leading to remarkable improvements in complex tasks. However, they pose great safety risks against harmful queries and adversarial…

Artificial Intelligence · Computer Science 2025-11-18 Kaiwen Zhou , Xuandong Zhao , Gaowen Liu , Jayanth Srinivasa , Aosong Feng , Dawn Song , Xin Eric Wang

Safety guardrails in large language models (LLMs) are a critical component in preventing harmful outputs. Yet, their resilience under perturbation remains poorly understood. In this paper, we investigate the robustness of safety fine-tuning…

Computation and Language · Computer Science 2025-10-14 Prithviraj Singh Shahani , Kaveh Eskandari Miandoab , Matthias Scheutz

This work introduces a novel framework for evaluating LLMs' capacity to balance instruction-following with critical reasoning when presented with multiple-choice questions containing no valid answers. Through systematic evaluation across…

Computation and Language · Computer Science 2025-06-03 Gracjan Góral , Emilia Wiśnios , Piotr Sankowski , Paweł Budzianowski

Large language models (LLMs) have made significant advancements across various tasks, but their safety alignment remain a major concern. Exploring jailbreak prompts can expose LLMs' vulnerabilities and guide efforts to secure them. Existing…

Cryptography and Security · Computer Science 2025-11-25 Xiaoning Dong , Wenbo Hu , Wei Xu , Tianxing He