paper-with-me

홈 › Papers

MM-PoisonRAG: Disrupting Multimodal RAG with Local and Global Poisoning Attacks

2025-02-25 · Hyeonjeong Ha, Qiusi Zhan, Jeonghwan Kim, Dimitrios Bralios, Saikrishna Sanniboina, Nanyun Peng, Kai-Wei Chang, Daniel Kang, Heng Ji

Multimodal large language models (MLLMs) equipped with Retrieval Augmented Generation (RAG) leverage both their rich parametric knowledge and the dynamic, external knowledge to excel in tasks such as Question Answering. While RAG enhances MLLMs by grounding responses in query-relevant external knowledge, this reliance poses a critical yet underexplored safety risk: knowledge poisoning attacks, where misinformation or irrelevant knowledge is intentionally injected into external knowledge bases to manipulate model outputs to be incorrect and even harmful. To expose such vulnerabilities in multimodal RAG, we propose MM-PoisonRAG, a novel knowledge poisoning attack framework with two attack strategies: Localized Poisoning Attack (LPA), which injects query-specific misinformation in both text and images for targeted manipulation, and Globalized Poisoning Attack (GPA) to provide false guidance during MLLM generation to elicit nonsensical responses across all queries. We evaluate our attacks across multiple tasks, models, and access settings, demonstrating that LPA successfully manipulates the MLLM to generate attacker-controlled answers, with a success rate of up to 56% on MultiModalQA. Moreover, GPA completely disrupts model generation to 0% accuracy with just a single irrelevant knowledge injection. Our results highlight the urgent need for robust defenses against knowledge poisoning to safeguard multimodal RAG frameworks.

📄 PDF Abstract BibTeX arXiv:2502.17832

Code (1)

hyeonjeongha/mm-poisonrag 공식 구현 pytorch

Tasks

MisinformationQuestion AnsweringRAGRetrieval-augmented Generation

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Weight Decay 설명 없음
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…
Attention Dropout Attention Dropout is a type of dropout used in attention-based architectures, where elements are randomly dropped out of the…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
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…
Residual Connection 설명 없음

Similar Papers 제목 키워드 기반

Hidden in the Metadata: Stealth Poisoning Attacks on Multimodal Retrieval-Augmented Generation

2026-02-26 · Kennedy Edemacu, Mohammad Mahdi Shokri arxiv

Retrieval-augmented generation (RAG) has emerged as a powerful paradigm for enhancing multimodal large language models by grounding their responses in external, factual knowledge and thus mitigating hallucinations. Howev…

No Vandalism: Privacy-Preserving and Byzantine-Robust Federated Learning

2024-06-03 · Zhibo Xing, Zijian Zhang, Zi'ang Zhang, Jiamou Liu 외

Federated learning allows several clients to train one machine learning model jointly without sharing private data, providing privacy protection. However, traditional federated learning is vulnerable to poisoning attacks…

Federated LearningModel PoisoningPrivacy Preserving

Performance Guaranteed Poisoning Attacks in Federated Learning: A Sliding Mode Approach

2025-05-22 · Huazi Pan, Yanjun Zhang, Leo Yu Zhang, Scott Adams 외

Manipulation of local training data and local updates, i.e., the poisoning attack, is the main threat arising from the collaborative nature of the federated learning (FL) paradigm. Most existing poisoning attacks aim to …

Federated LearningModel Poisoning

Dual Defense: Enhancing Privacy and Mitigating Poisoning Attacks in Federated Learning

2025-02-08 · Runhua Xu, Shiqi Gao, Chao Li, James Joshi 외

Federated learning (FL) is inherently susceptible to privacy breaches and poisoning attacks. To tackle these challenges, researchers have separately devised secure aggregation mechanisms to protect data privacy and robus…

Anomaly DetectionFederated LearningModel Poisoning

Local Model Poisoning Attacks to Byzantine-Robust Federated Learning

2019-11-26 · Minghong Fang, Xiaoyu Cao, Jinyuan Jia, Neil Zhenqiang Gong

In federated learning, multiple client devices jointly learn a machine learning model: each client device maintains a local model for its local training dataset, while a master device maintains a global model via aggrega…

BIG-bench Machine LearningData PoisoningFederated LearningModel Poisoning