English
Related papers

Related papers: A Granular Study of Safety Pretraining under Model…

200 papers

Large Language Models (LLMs) can comply with harmful instructions, raising serious safety concerns despite their impressive capabilities. Recent work has leveraged probing-based approaches to study the separability of malicious and benign…

Computation and Language · Computer Science 2025-12-16 Cheng Wang , Zeming Wei , Qin Liu , Muhao Chen

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

Video generative models achieve high-quality synthesis from natural-language prompts by leveraging large-scale web data. However, this training paradigm inherently exposes them to unsafe biases and harmful concepts, introducing the risk of…

Modern large language models (LLMs) are often evaluated and deployed under a one-shot, greedy inference protocol, especially in professional settings that require deterministic behavior. This regime can systematically under-estimate a fixed…

Artificial Intelligence · Computer Science 2026-02-13 Xinhai Sun

Large language models (LLMs) are increasingly deployed behind safety guardrails such as system prompts and content filters, especially in settings where product teams cannot modify model weights. In practice these guardrails are typically…

Cryptography and Security · Computer Science 2025-12-19 Perry Abdulkadir

Large language models (LLMs) are increasingly used for decision-making tasks under uncertainty; however, their risk profiles and how they are influenced by prompting and alignment methods remain underexplored. Existing studies have…

Artificial Intelligence · Computer Science 2025-10-08 Yikai Wang , Xiaocheng Li , Guanting Chen

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

The widespread popularity of Large Language Models (LLMs), partly due to their unique ability to perform in-context learning, has also brought to light the importance of ethical and safety considerations when deploying these pre-trained…

Computation and Language · Computer Science 2024-08-07 Karuna Bhaila , Minh-Hao Van , Xintao Wu

Prompt design significantly impacts the moral competence and safety alignment of large language models (LLMs), yet empirical comparisons remain fragmented across datasets and models.We introduce ProMoral-Bench, a unified benchmark…

Artificial Intelligence · Computer Science 2026-02-17 Rohan Subramanian Thomas , Shikhar Shiromani , Abdullah Chaudhry , Ruizhe Li , Vasu Sharma , Kevin Zhu , Sunishchal Dev

Safely aligning large language models (LLMs) often demands extensive human-labeled preference data, a process that's both costly and time-consuming. While synthetic data offers a promising alternative, current methods frequently rely on…

Cryptography and Security · Computer Science 2025-06-13 Kyubyung Chae , Hyunbin Jin , Taesup Kim

Large language models (LLMs) show inherent brittleness in their safety mechanisms, as evidenced by their susceptibility to jailbreaking and even non-malicious fine-tuning. This study explores this brittleness of safety alignment by…

Machine Learning · Computer Science 2024-10-28 Boyi Wei , Kaixuan Huang , Yangsibo Huang , Tinghao Xie , Xiangyu Qi , Mengzhou Xia , Prateek Mittal , Mengdi Wang , Peter Henderson

Safe reinforcement learning (RL) agents accomplish given tasks while adhering to specific constraints. Employing constraints expressed via easily-understandable human language offers considerable potential for real-world applications due to…

Machine Learning · Computer Science 2024-05-16 Xingzhou Lou , Junge Zhang , Ziyan Wang , Kaiqi Huang , Yali Du

Language models, characterized by their black-box nature, often hallucinate and display sensitivity to input perturbations, causing concerns about trust. To enhance trust, it is imperative to gain a comprehensive understanding of the…

Computation and Language · Computer Science 2025-01-03 Vatsal Gupta , Pranshu Pandya , Tushar Kataria , Vivek Gupta , Dan Roth

Open-weight models provide researchers and developers with accessible foundations for diverse downstream applications. We tested the safety and security postures of eight open-weight large language models (LLMs) to identify vulnerabilities…

Cryptography and Security · Computer Science 2025-11-06 Amy Chang , Nicholas Conley , Harish Santhanalakshmi Ganesan , Adam Swanda

Safety evaluations of large language models (LLMs) typically report binary outcomes, i.e. attack success rate (ASR), refusal rate, or harmful versus safe classification, which hide how risk changes between prompt and response. We present a…

Computation and Language · Computer Science 2026-05-21 Mengya Hu , Qiong Wei , Sandeep Atluri

Large Language Models (LLMs) often produce explanations that do not faithfully reflect the factors driving their predictions. In healthcare settings, such unfaithfulness is especially problematic: explanations that omit salient clinical…

Computation and Language · Computer Science 2025-11-04 Teague McMillan , Gabriele Dominici , Martin Gjoreski , Marc Langheinrich

In this paper, we investigate the degree to which fine-tuning in Large Language Models (LLMs) effectively mitigates versus merely conceals undesirable behavior. Through the lens of semi-realistic role-playing exercises designed to elicit…

Computation and Language · Computer Science 2024-07-01 Florin Pop , Judd Rosenblatt , Diogo Schwerz de Lucena , Michael Vaiana

Large language models (LLMs) have revolutionized numerous domains with their impressive performance but still face their challenges. A predominant issue is the propensity for these models to generate non-existent facts, a concern termed…

Computation and Language · Computer Science 2024-06-10 Hanning Zhang , Shizhe Diao , Yong Lin , Yi R. Fung , Qing Lian , Xingyao Wang , Yangyi Chen , Heng Ji , Tong Zhang

While metrics available during pre-training, such as perplexity, correlate well with model performance at scaling-laws studies, their predictive capacities at a fixed model size remain unclear, hindering effective model selection and…

Computation and Language · Computer Science 2025-10-17 Hansi Zeng , Kai Hui , Honglei Zhuang , Zhen Qin , Zhenrui Yue , Hamed Zamani , Dana Alon

Activation steering is a practical post-training model alignment technique to enhance the utility of Large Language Models (LLMs). Prior to deploying a model as a service, developers can steer a pre-trained model toward specific behavioral…

Cryptography and Security · Computer Science 2026-02-06 Chen Xiong , Zhiyuan He , Pin-Yu Chen , Ching-Yun Ko , Tsung-Yi Ho