中文
相关论文

相关论文: Can Safety Fine-Tuning Be More Principled? Lessons…

200 篇论文

The safety alignment of current Large Language Models (LLMs) is vulnerable. Relatively simple attacks, or even benign fine-tuning, can jailbreak aligned models. We argue that many of these vulnerabilities are related to a shared underlying…

密码学与安全 · 计算机科学 2024-06-11 Xiangyu Qi , Ashwinee Panda , Kaifeng Lyu , Xiao Ma , Subhrajit Roy , Ahmad Beirami , Prateek Mittal , Peter Henderson

Large Language Models (LLMs) are increasingly popular, powering a wide range of applications. Their widespread use has sparked concerns, especially through jailbreak attacks that bypass safety measures to produce harmful content. In this…

密码学与安全 · 计算机科学 2025-12-25 Zhengchun Shang , Wenlan Wei , Weiheng Bai

Caution: This paper includes offensive words that could potentially cause unpleasantness. Language models (LMs) are vulnerable to exploitation for adversarial misuse. Training LMs for safety alignment is extensive and makes it hard to…

机器学习 · 计算机科学 2024-02-28 Heegyu Kim , Sehyun Yuk , Hyunsouk Cho

As large language models (LLMs) become ubiquitous, parameter-efficient fine-tuning methods and safety-first defenses have proliferated rapidly. However, the number of approaches and their recent increase have resulted in diverse…

机器学习 · 计算机科学 2025-06-03 Saad Hossain , Samanvay Vajpayee , Sirisha Rambhatla

With the rapid advancement of large language models (LLMs), ensuring their safe use becomes increasingly critical. Fine-tuning is a widely used method for adapting models to downstream tasks, yet it is vulnerable to jailbreak attacks.…

密码学与安全 · 计算机科学 2025-10-10 Xiangfang Li , Yu Wang , Bo Li

The AI era has ushered in Large Language Models (LLM) to the technological forefront, which has been much of the talk in 2023, and is likely to remain as such for many years to come. LLMs are the AI models that are the power house behind…

密码学与安全 · 计算机科学 2026-01-22 Anjanava Biswas , Wrick Talukdar

Jailbreaking is an emerging adversarial attack that bypasses the safety alignment deployed in off-the-shelf large language models (LLMs) and has evolved into multiple categories: human-based, optimization-based, generation-based, and the…

密码学与安全 · 计算机科学 2025-02-06 Xunguang Wang , Daoyuan Wu , Zhenlan Ji , Zongjie Li , Pingchuan Ma , Shuai Wang , Yingjiu Li , Yang Liu , Ning Liu , Juergen Rahmel

With the recent proliferation of large language models (LLMs), enterprises have been able to rapidly develop proof-of-concepts and prototypes. As a result, there is a growing need to implement robust guardrails that monitor, quantize and…

计算与语言 · 计算机科学 2025-10-20 Aaron Zheng , Mansi Rana , Andreas Stolcke

The jailbreak attack can bypass the safety measures of a Large Language Model (LLM), generating harmful content. This misuse of LLM has led to negative societal consequences. Currently, there are two main approaches to address jailbreak…

计算与语言 · 计算机科学 2024-03-25 Zezhong Wang , Fangkai Yang , Lu Wang , Pu Zhao , Hongru Wang , Liang Chen , Qingwei Lin , Kam-Fai Wong

Large Language Models (LLMs) are vulnerable to jailbreak attacks that exploit weaknesses in traditional safety alignment, which often relies on rigid refusal heuristics or representation engineering to block harmful outputs. While they are…

计算与语言 · 计算机科学 2025-10-01 Yuyou Zhang , Miao Li , William Han , Yihang Yao , Zhepeng Cen , Ding Zhao

Although safely enhanced Large Language Models (LLMs) have achieved remarkable success in tackling various complex tasks in a zero-shot manner, they remain susceptible to jailbreak attacks, particularly the unknown jailbreak attack. To…

计算与语言 · 计算机科学 2024-06-12 Fan Liu , Zhao Xu , Hao Liu

LLMs(Large Language Models) nowadays have widespread adoption as a tool for solving issues across various domain/tasks. These models since are susceptible to produce harmful or toxic results, inference-time adversarial attacks, therefore…

计算与语言 · 计算机科学 2024-10-01 Sachin Kumar

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…

计算与语言 · 计算机科学 2024-10-15 Minjun Zhu , Linyi Yang , Yifan Wei , Ningyu Zhang , Yue Zhang

The integration of large language models (LLMs) into cyber security applications presents both opportunities and critical safety risks. We introduce CyberLLMInstruct, a dataset of 54,928 pseudo-malicious instruction-response pairs spanning…

密码学与安全 · 计算机科学 2025-09-18 Adel ElZemity , Budi Arief , Shujun Li

Stakeholders -- from model developers to policymakers -- seek to minimize the dual-use risks of large language models (LLMs). An open challenge to this goal is whether technical safeguards can impede the misuse of LLMs, even when models are…

This paper introduces MetaDefense, a novel framework for defending against finetuning-based jailbreak attacks in large language models (LLMs). We observe that existing defense mechanisms fail to generalize to harmful queries disguised by…

机器学习 · 计算机科学 2025-10-10 Weisen Jiang , Sinno Jialin Pan

Large language models (LLMs) have emerged as powerful tools for addressing a wide range of general inquiries and tasks. Despite this, fine-tuning aligned LLMs on smaller, domain-specific datasets, critical to adapting them to specialized…

人工智能 · 计算机科学 2025-02-04 Guanlin Li , Kangjie Chen , Shangwei Guo , Jie Zhang , Han Qiu , Chao Zhang , Guoyin Wang , Tianwei Zhang , Jiwei Li

Large Language Models (LLMs), which bridge the gap between human language understanding and complex problem-solving, achieve state-of-the-art performance on several NLP tasks, particularly in few-shot and zero-shot settings. Despite the…

密码学与安全 · 计算机科学 2025-01-07 Shuai Zhao , Meihuizi Jia , Zhongliang Guo , Leilei Gan , Xiaoyu Xu , Xiaobao Wu , Jie Fu , Yichao Feng , Fengjun Pan , Luu Anh Tuan

Recent defenses for safeguarding open-weight large language models (LLMs) are intended to prevent adversarial usage. Underlying these defenses is an assumption that new harmful behavior is learned through fine-tuning rather than elicited by…

机器学习 · 计算机科学 2026-05-27 Kevin Kuo , Chhavi Yadav , Virginia Smith

Machine learning (ML) models deployed in many safety- and business-critical systems are vulnerable to exploitation through adversarial examples. A large body of academic research has thoroughly explored the causes of these blind spots,…

密码学与安全 · 计算机科学 2020-07-15 Ivan Evtimov , Weidong Cui , Ece Kamar , Emre Kiciman , Tadayoshi Kohno , Jerry Li