Vis-Poison: Poisoning Visual Knowledge in Multimodal Retrieval-Augmented Generation

arXiv:2608.20756 · cs.CV, cs.AI · Submitted 2026-08-21 · Read on arXiv

cs.CV, cs.AI

Submitted: 2026-08-21

Updated: 2026-08-28

Comments: Findings of EMNLP, 2026

Code: https://github.com/SWUFE-DB-Group/Vis-Poison

Project page: https://openai.github.io/openai-guardrails-python/ref/checks/jailbreak

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• RAG — 45× in this paper · explained in 3 episodes
• Alignment — 7× in this paper · explained in 1 episode
• Verification — 7× in this paper · explained in 2 episodes
• Verifier — 4× in this paper · explained in 2 episodes
• Retrieval-Augmented Generation (RAG) — 3× in this paper · explained in 16 episodes
• Robustness — 3× in this paper · explained in 5 episodes
• Vision-language models — 3× in this paper · explained in 2 episodes
• Benchmark — 2× in this paper · explained in 2 episodes
• Consistency — 2× in this paper · explained in 3 episodes
• Embeddings — 2× in this paper · explained in 2 episodes
• Semantic Alignment — 2× in this paper · explained in 3 episodes
• Accuracy — 1× in this paper · explained in 1 episode
• Adversarial Attacks — 1× in this paper · explained in 1 episode
• Clean Help Rate — 1× in this paper
• Few-Shot Learning — 1× in this paper · explained in 4 episodes
• Foundation Model — 1× in this paper · explained in 3 episodes
• Foundation Models — 1× in this paper · explained in 7 episodes
• Hallucination — 1× in this paper · explained in 4 episodes
• Instruction Tuning — 1× in this paper · explained in 2 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 6 episodes
• Model-Agnostic — 1× in this paper · explained in 2 episodes
• Multimodal Inputs — 1× in this paper · explained in 2 episodes
• Poison Induction Rate — 1× in this paper
• Poison Override Rate — 1× in this paper
• Steering — 1× in this paper · explained in 2 episodes
• Structural Reasoning — 1× in this paper · explained in 2 episodes
• Transferability — 1× in this paper · explained in 4 episodes

Sources

• SegSub: Evaluating Robustness to Knowledge Conflicts and Hallucinations in Vision-Language Models
• Hidden in the Metadata: Stealth Poisoning Attacks on Multimodal Retrieval-Augmented Generation
• Qwen3-VL Technical Report
• MM-PoisonRAG: Disrupting Multimodal RAG with Local and Global Poisoning Attacks
• Beyond Text: Optimizing RAG with Multimodal Inputs for Industrial Applications
• CPA-RAG:Covert Poisoning Attacks on Retrieval-Augmented Generation in Large Language Models
• Adversarial attacks against Modern Vision-Language Models
• Spa-VLM: Stealthy Poisoning Attacks on RAG-based VLM
• AVA-Bench: Atomic Visual Ability Benchmark for Vision Foundation Models

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models