paper-with-me

Papers

Alphabet Index Mapping: Jailbreaking LLMs through Semantic Dissimilarity

2025-06-15 · Bilal Saleh Husain

Large Language Models (LLMs) have demonstrated remarkable capabilities, yet their susceptibility to adversarial attacks, particularly jailbreaking, poses significant safety and ethical concerns. While numerous jailbreak methods exist, many suffer from computational expense, high token usage, or complex decoding schemes. Liu et al. (2024) introduced FlipAttack, a black-box method that achieves high attack success rates (ASR) through simple prompt manipulation. This paper investigates the underlying mechanisms of FlipAttack's effectiveness by analyzing the semantic changes induced by its flipping modes. We hypothesize that semantic dissimilarity between original and manipulated prompts is inversely correlated with ASR. To test this, we examine embedding space visualizations (UMAP, KDE) and cosine similarities for FlipAttack's modes. Furthermore, we introduce a novel adversarial attack, Alphabet Index Mapping (AIM), designed to maximize semantic dissimilarity while maintaining simple decodability. Experiments on GPT-4 using a subset of AdvBench show AIM and its variant AIM+FWO achieve a 94% ASR, outperforming FlipAttack and other methods on this subset. Our findings suggest that while high semantic dissimilarity is crucial, a balance with decoding simplicity is key for successful jailbreaking. This work contributes to a deeper understanding of adversarial prompt mechanics and offers a new, effective jailbreak technique.

📄 PDF Abstract BibTeX arXiv:2506.12685

Code (0)

등록된 구현이 없습니다.

Tasks

Adversarial Attack

Methods 이 논문이 사용한 방법론

Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Transformer A Transformer is a model architecture that eschews recurrence and instead relies entirely on an [attention…

Similar Papers 제목 키워드 기반

AutoBreach: Universal and Adaptive Jailbreaking with Efficient Wordplay-Guided Optimization

2024-05-30 · Jiawei Chen, Xiao Yang, Zhengwei Fang, Yu Tian 외

Despite the widespread application of large language models (LLMs) across various tasks, recent studies indicate that they are susceptible to jailbreak attacks, which can render their defense mechanisms ineffective. Howe…

SentenceSentence Compression

Safe2Harm: Semantic Isomorphism Attacks for Jailbreaking Large Language Models

2025-12-05 · Fan Yang arxiv

Large Language Models (LLMs) have demonstrated exceptional performance across various tasks, but their security vulnerabilities can be exploited by attackers to generate harmful content, causing adverse impacts across va…

Prompt Engineering

Exploiting the Index Gradients for Optimization-Based Jailbreaking on Large Language Models

2024-12-11 · Jiahui Li, Yongchang Hao, Haoyu Xu, Xing Wang 외

Despite the advancements in training Large Language Models (LLMs) with alignment techniques to enhance the safety of generated content, these models remain susceptible to jailbreak, an adversarial attack method that expo…

Adversarial Attack

Jailbreaking Large Language Models Through Alignment Vulnerabilities in Out-of-Distribution Settings

2024-06-19 · Yue Huang, Jingyu Tang, Dongping Chen, Bingda Tang 외

Recently, Large Language Models (LLMs) have garnered significant attention for their exceptional natural language processing capabilities. However, concerns about their trustworthiness remain unresolved, particularly in …

Multi-turn Jailbreaking Attack in Multi-Modal Large Language Models

2026-01-08 · Badhan Chandra Das, Md Tasnim Jawad, Joaquin Molto, M. Hadi Amini 외 arxiv

In recent years, the security vulnerabilities of Multi-modal Large Language Models (MLLMs) have become a serious concern in the Generative Artificial Intelligence (GenAI) research. These highly intelligent models, capabl…