English
Related papers

Related papers: BioRefusalAudit: Auditing Biosecurity Refusal Dept…

200 papers

This study evaluates the performance of Large Language Models (LLMs) on SemEval-2020 Task 4 dataset, focusing on commonsense validation and explanation. Our methodology involves evaluating multiple LLMs, including LLaMA3-70B, Gemma2-9B, and…

Computation and Language · Computer Science 2025-02-25 Rawand Alfugaha , Mohammad AL-Smadi

How safety supervision is written may matter more than the explicit identity content it contains. We study low-data LoRA safety fine-tuning with four supervision formats built from the same core safety rules: constitutional rules (A),…

Computation and Language · Computer Science 2026-03-17 Xinran Zhang

Large language models can be uncertain yet correct, or confident yet wrong, raising the question of whether their output-level uncertainty and their actual correctness are driven by the same internal mechanisms or by distinct feature…

Machine Learning · Computer Science 2026-04-23 Het Patel , Tiejin Chen , Hua Wei , Evangelos E. Papalexakis , Jia Chen

Backdoor mechanisms have traditionally been studied as security threats that compromise the integrity of machine learning models. However, the same mechanism -- the conditional activation of specific behaviors through input triggers -- can…

Cryptography and Security · Computer Science 2026-03-10 Yige Li , Wei Zhao , Zhe Li , Nay Myat Min , Hanxun Huang , Yunhan Zhao , Xingjun Ma , Yu-Gang Jiang , Jun Sun

Prior work argues that refusal in large language models is mediated by a single activation-space direction, enabling effective steering and ablation. We show that this account is incomplete. Across eleven categories of refusal and…

Computation and Language · Computer Science 2026-02-03 Faaiz Joad , Majd Hawasly , Sabri Boughorbel , Nadir Durrani , Husrev Taha Sencar

The emergence of Large Language Models (LLMs) has significantly influenced various aspects of software development activities. Despite their benefits, LLMs also pose notable risks, including the potential to generate harmful content and…

Software Engineering · Computer Science 2024-09-24 Jiachi Chen , Qingyuan Zhong , Yanlin Wang , Kaiwen Ning , Yongkun Liu , Zenan Xu , Zhe Zhao , Ting Chen , Zibin Zheng

Large Language models (LLMs) have demonstrated state-of-the-art performance in various natural language processing (NLP) tasks across multiple domains, yet they are prone to shortcut learning and factual inconsistencies. This research…

Computation and Language · Computer Science 2024-04-09 Shreyasi Mandal , Ashutosh Modi

The rapid advancement of large language models (LLMs) has brought significant benefits to various domains while introducing substantial risks. Despite being fine-tuned through reinforcement learning, LLMs lack the capability to discern…

Computation and Language · Computer Science 2025-04-15 Weilong Dong , Peiguang Li , Yu Tian , Xinyi Zeng , Fengdi Li , Sirui Wang

LLMs increasingly exhibit over-refusal behavior, where safety mechanisms cause models to reject benign instructions that seemingly resemble harmful content. This phenomenon diminishes utility in production applications that repeatedly rely…

Computation and Language · Computer Science 2026-04-21 Utsav Maskey , Sumit Yadav , Mark Dras , Usman Naseem

Recent studies have widely investigated backdoor attacks on Large Language Models (LLMs) by inserting harmful question-answer (QA) pairs into their training data. However, we revisit existing attacks and identify two critical limitations:…

Computation and Language · Computer Science 2025-10-07 Jiawei Kong , Hao Fang , Xiaochen Yang , Kuofeng Gao , Bin Chen , Shu-Tao Xia , Ke Xu , Han Qiu

Multilingual large language models (LLMs) exhibit strong cross-linguistic generalization, yet medium to low resource languages underperform on common benchmarks such as ARC-Challenge, MMLU, and HellaSwag. We analyze activation patterns in…

Computation and Language · Computer Science 2025-07-28 Richmond Sin Jing Xuan , Jalil Huseynov , Yang Zhang

Large language models are meticulously aligned to be both helpful and harmless. However, recent research points to a potential overkill which means models may refuse to answer benign queries. In this paper, we investigate the factors for…

Computation and Language · Computer Science 2024-02-01 Chenyu Shi , Xiao Wang , Qiming Ge , Songyang Gao , Xianjun Yang , Tao Gui , Qi Zhang , Xuanjing Huang , Xun Zhao , Dahua Lin

Large Language Models (LLMs) can be tasked with scoring texts according to pre-defined criteria and on a defined scale, but there is no recognised optimal prompting strategy for this. This article focuses on the task of LLMs scoring journal…

Digital Libraries · Computer Science 2025-12-02 Mike Thelwall

When the substantive content of a request is rewritten, do large language models still answer in the format the original task asked for? We find that they often do not, even at temperature zero. On a 150-query evaluation over five compact…

Computation and Language · Computer Science 2026-05-12 Aofan Liu , Jingxiang Meng

Safety-aligned language models must refuse harmful requests without broad over-refusal, but it remains unclear how dynamic adversarial fine-tuning changes refusal-control carriers: Kullback--Leibler (KL)-constrained directions or small…

Machine Learning · Computer Science 2026-05-27 Wenhao Lan , Shan Li , Xinhua Lai , Meiqi Wu , Junbin Yang , Haihua Shen , Yijun Yang

Large Language Models (LLMs) such as Gemma-2B have shown strong performance in various natural language processing tasks. However, general-purpose models often lack the domain expertise required for cybersecurity applications. This work…

Cryptography and Security · Computer Science 2026-01-13 Vasanth Iyer , Leonardo Bobadilla , S. S. Iyengar

Safety alignment -- training large language models (LLMs) to refuse harmful requests while remaining helpful -- is critical for responsible deployment. Prior work established that safety behaviors are governed by low-rank structures,…

Computation and Language · Computer Science 2026-01-06 Dianyun Wang , Qingsen Ma , Yuhu Shang , Zhifeng Lu , Zhenbo Xu , Lechen Ning , Huijia Wu , Zhaofeng He

While prior work has focused on projecting adversarial examples back onto the manifold of natural data to restore safety, we argue that a comprehensive understanding of AI safety requires characterizing the unsafe regions themselves. This…

Machine Learning · Computer Science 2026-05-07 Sarthak Munshi , Manish Bhatt , Vineeth Sai Narajala , Idan Habler , Ammar Al-Kahfah , Ken Huang , Blake Gatto

Multi-turn prompt injection follows a known attack path -- trust-building, pivoting, escalation but text-level defenses miss covert attacks where individual turns appear benign. We show this attack path leaves an activation-level signature…

Cryptography and Security · Computer Science 2026-05-01 Prashant Kulkarni

Most commonly used language models (LMs) are instruction-tuned and aligned using a combination of fine-tuning and reinforcement learning, causing them to refuse users requests deemed harmful by the model. However, jailbreak prompts can…

Computation and Language · Computer Science 2025-07-02 Aryan Shrivastava , Ari Holtzman
‹ Prev 1 8 9 10 Next ›