English
Related papers

Related papers: When Context Flips, Safety Breaks: Diagnosing Brit…

200 papers

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

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

Machine learning models are often brittle on production data despite achieving high accuracy on benchmark datasets. Benchmark datasets have traditionally served dual purposes: first, benchmarks offer a standard on which machine learning…

Machine Learning · Computer Science 2022-09-26 Matthew Groh

As Large Language Models (LLMs) continue to advance in understanding and generating long sequences, new safety concerns have been introduced through the long context. However, the safety of LLMs in long-context tasks remains under-explored,…

Computation and Language · Computer Science 2025-02-25 Yida Lu , Jiale Cheng , Zhexin Zhang , Shiyao Cui , Cunxiang Wang , Xiaotao Gu , Yuxiao Dong , Jie Tang , Hongning Wang , Minlie Huang

While Large Language Models (LLMs) are widely documented to be sensitive to minor prompt perturbations and prone to sycophantic alignment, their robustness in consequential, rule-bound decision-making remains under-explored. We uncover a…

Artificial Intelligence · Computer Science 2026-04-07 Jon Chun , Katherine Elkins

As vision-language models (VLMs) become increasingly capable, maintaining a balance between safety and usefulness remains a central challenge. Safety mechanisms, while essential, can backfire, causing over-refusal, where models decline…

Computation and Language · Computer Science 2026-03-20 Kaixuan Ren , Preslav Nakov , Usman Naseem

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

The robustness of machine learning models can be compromised by spurious correlations between non-causal features in the input data and target labels. A common way to test for such correlations is to train on data where the label is…

Computer Vision and Pattern Recognition · Computer Science 2026-04-16 Akshit Achara , Yovin Yathathugoda , Nick Byrne , Michela Antonelli , Esther Puyol Anton , Alexander Hammers , Andrew P. King

Large language models (LLMs) have advanced the development of personalized learning in education. However, their inherent generation mechanisms often produce homogeneous responses to identical prompts. This one-size-fits-all mechanism…

Computation and Language · Computer Science 2026-02-06 Rui Jia , Ruiyi Lan , Fengrui Liu , Zhongxiang Dai , Bo Jiang , Jing Shao , Jingyuan Chen , Guandong Xu , Fei Wu , Min Zhang

Instruction-following language models are trained to be helpful and safe, yet their safety behavior can deteriorate under benign fine-tuning and worsen under adversarial updates. Existing defenses often offer limited protection or force a…

Computation and Language · Computer Science 2026-05-12 Jyotin Goel , Souvik Maji , Pratik Mazumder

Recent advancements in Large Language Models (LLMs) have sparked widespread concerns about their safety. Recent work demonstrates that safety alignment of LLMs can be easily removed by fine-tuning with a few adversarially chosen…

Computation and Language · Computer Science 2025-03-03 Samuele Poppi , Zheng-Xin Yong , Yifei He , Bobbie Chern , Han Zhao , Aobo Yang , Jianfeng Chi

Recent advancements in Large Language Models (LLMs) have significantly enhanced interactions between users and models. These advancements concurrently underscore the need for rigorous safety evaluations due to the manifestation of social…

Computation and Language · Computer Science 2025-03-26 Dahyun Jung , Seungyoon Lee , Hyeonseok Moon , Chanjun Park , Heuiseok Lim

We survey the state-of-the-art on model-based formalisms for safety and security joint analysis, where safety refers to the absence of unintended failures, and security to absence of malicious attacks. We conduct a thorough literature…

Cryptography and Security · Computer Science 2023-10-24 Stefano M. Nicoletti , Marijn Peppelman , Christina Kolb , Mariëlle Stoelinga

As Large Language Models transition to autonomous agents, user inputs frequently violate cooperative assumptions (e.g., implicit intent, missing parameters, false presuppositions, or ambiguous expressions), creating execution risks that…

Artificial Intelligence · Computer Science 2026-02-03 Han Bao , Zheyuan Zhang , Pengcheng Jing , Zhengqing Yuan , Kaiwen Shi , Yanfang Ye

Safety alignment is a key requirement for building reliable Artificial General Intelligence. Despite significant advances in safety alignment, we observe that minor latent shifts can still trigger unsafe responses in aligned models. We…

Machine Learning · Computer Science 2025-06-23 Tianle Gu , Kexin Huang , Zongqi Wang , Yixu Wang , Jie Li , Yuanqi Yao , Yang Yao , Yujiu Yang , Yan Teng , Yingchun Wang

Large language models require continuous adaptation to new tasks while preserving safety alignment. However, fine-tuning on even benign data often compromises safety behaviors, including refusal of harmful requests, truthfulness, and…

Machine Learning · Computer Science 2026-04-21 Thong Bach , Dung Nguyen , Thao Minh Le , Truyen Tran

General Alignment has improved average-case helpfulness and safety, but current alignment practice still rewards confident, single-turn responses. The problem is not only that models fail on edge cases; it is that current evaluation makes…

Computation and Language · Computer Science 2026-05-19 Han Bao , Yue Huang , Xiaoda Wang , Zheyuan Zhang , Yujun Zhou , Carl Yang , Xiangliang Zhang , Yanfang Ye

A standard practice when using large language models is for users to supplement their instruction with an input context containing new information for the model to process. However, models struggle to reliably follow the input context,…

Machine Learning · Computer Science 2025-04-22 Sachin Goyal , Christina Baek , J. Zico Kolter , Aditi Raghunathan

Consider a scenario where a harmfulness evaluation metric intended to filter unsafe responses from a Large Language Model. When applied to individual harmful prompt-response pairs, it correctly flags them as unsafe by assigning a high-risk…

Artificial Intelligence · Computer Science 2025-02-14 Ora Nova Fandina , Leshem Choshen , Eitan Farchi , George Kour , Yotam Perlitz , Orna Raz

Reusable skills are becoming a common interface for extending large language model agents, packaging procedural guidance with access to files, tools, memory, and execution environments. However, this modularity introduces attack surfaces…

Cryptography and Security · Computer Science 2026-05-28 Chang Jin , An Wang , Zeming Wei , Kai Wang , Biaojie Zeng , Qiaosheng Zhang , Chao Yang , Jingjing Qu , Xia Hu , Xingcheng Xu