paper-with-me

Papers

"Not Aligned" is Not "Malicious": Being Careful about Hallucinations of Large Language Models' Jailbreak

2024-06-17 · Lingrui Mei, Shenghua Liu, Yiwei Wang, Baolong Bi, Jiayi Mao, Xueqi Cheng

"Jailbreak" is a major safety concern of Large Language Models (LLMs), which occurs when malicious prompts lead LLMs to produce harmful outputs, raising issues about the reliability and safety of LLMs. Therefore, an effective evaluation of jailbreaks is very crucial to develop its mitigation strategies. However, our research reveals that many jailbreaks identified by current evaluations may actually be hallucinations-erroneous outputs that are mistaken for genuine safety breaches. This finding suggests that some perceived vulnerabilities might not represent actual threats, indicating a need for more precise red teaming benchmarks. To address this problem, we propose the $\textbf{B}$enchmark for reli$\textbf{AB}$ilit$\textbf{Y}$ and jail$\textbf{B}$reak ha$\textbf{L}$l$\textbf{U}$cination $\textbf{E}$valuation (BabyBLUE). BabyBLUE introduces a specialized validation framework including various evaluators to enhance existing jailbreak benchmarks, ensuring outputs are useful malicious instructions. Additionally, BabyBLUE presents a new dataset as an augmentation to the existing red teaming benchmarks, specifically addressing hallucinations in jailbreaks, aiming to evaluate the true potential of jailbroken LLM outputs to cause harm to human society.

📄 PDF Abstract BibTeX arXiv:2406.11668

Code (1)

Meirtz/BabyBLUE-llm 공식 구현

Tasks

Red Teaming

Similar Papers 제목 키워드 기반

EnJa: Ensemble Jailbreak on Large Language Models

2024-08-07 · Jiahao Zhang, Zilong Wang, Ruofan Wang, Xingjun Ma 외

As Large Language Models (LLMs) are increasingly being deployed in safety-critical applications, their vulnerability to potential jailbreaks -- malicious prompts that can disable the safety mechanism of LLMs -- has attra…

Safety Alignment

Fine-grained Hallucination Detection and Editing for Language Models

2024-01-12 · Abhika Mishra, Akari Asai, Vidhisha Balachandran, Yizhong Wang 외

Large language models (LMs) are prone to generate factual errors, which are often called hallucinations. In this paper, we introduce a comprehensive taxonomy of hallucinations and argue that hallucinations manifest in di…

HallucinationRetrieval

Transferable Ensemble Black-box Jailbreak Attacks on Large Language Models

2024-10-31 · Yiqi Yang, Hongye Fu

In this report, we propose a novel black-box jailbreak attacking framework that incorporates various LLM-as-Attacker methods to deliver transferable and powerful jailbreak attacks. Our method is designed based on three k…

A Multi-Dimensional Audit of Politically Aligned Large Language Models

2026-04-27 · Lisa Korver, Mohamed Mostagir, Sherief Reda arxiv

As the application of Large Language Models (LLMs) spreads across various industries, there are increasing concerns about the potential for their misuse, especially in sensitive areas such as political discourse. Deliber…

Prompt Engineering

Fine-Grained Detection of Context-Grounded Hallucinations Using LLMs

2025-09-26 · Yehonatan Peisakhovsky, Zorik Gekhman, Yosi Mass, Liat Ein-Dor 외 arxiv

Context-grounded hallucinations are cases where model outputs contain information not verifiable against the source text. We study the applicability of LLMs for localizing such hallucinations, as a more practical alterna…