FENCE: A Financial and Multimodal Jailbreak Detection Dataset

arXiv:2602.18154 · cs.CL, cs.AI, cs.DB · Submitted 2026-02-20 · Read on arXiv

cs.CL, cs.AI, cs.DB

Submitted: 2026-02-20

Updated: 2026-08-28

Code: https://github.com/kakaobank/FENCE

Terminology

• Benchmark — 26× in this paper · explained in 2 episodes
• Robustness — 8× in this paper · explained in 5 episodes
• Accuracy — 7× in this paper · explained in 1 episode
• Alignment — 7× in this paper · explained in 1 episode
• Fine-Tuning — 5× in this paper · explained in 5 episodes
• Adversarial Attacks — 4× in this paper · explained in 1 episode
• Attack Success Rate — 4× in this paper
• Generalization — 4× in this paper · explained in 13 episodes
• Image-based Attacks — 3× in this paper
• Multimodal Dataset — 3× in this paper · explained in 2 episodes
• Text-based Attacks — 3× in this paper
• Verification — 3× in this paper · explained in 2 episodes
• Adversarial Examples — 2× in this paper · explained in 2 episodes
• Large Language Models (LLMs) — 2× in this paper · explained in 6 episodes
• Prompt Engineering — 2× in this paper · explained in 4 episodes
• Adam — 1× in this paper · explained in 2 episodes
• Cosine Similarity — 1× in this paper · explained in 3 episodes
• Data Augmentation — 1× in this paper · explained in 7 episodes
• Defense Success Rate — 1× in this paper
• Distillation — 1× in this paper · explained in 2 episodes
• Domain Shift — 1× in this paper · explained in 4 episodes
• Embeddings — 1× in this paper · explained in 2 episodes
• Knowledge Distillation — 1× in this paper · explained in 11 episodes
• Multimodal Inputs — 1× in this paper · explained in 2 episodes
• Semantic Alignment — 1× in this paper · explained in 3 episodes
• Semantic Similarity — 1× in this paper · explained in 2 episodes

Sources

• GPT-4 Technical Report
• SemEval-2017 Task 1: Semantic Textual Similarity - Multilingual and Cross-lingual Focused Evaluation
• Llama Guard 3 Vision: Safeguarding Human-AI Image Understanding Conversations
• FigStep: Jailbreaking Large Vision-Language Models via Typographic Visual Prompts
• A Survey of Attacks on Large Vision-Language Models: Resources, Advances, and Future Trends
• Prompt Injection attack against LLM-integrated Applications
• Jailbreaking ChatGPT via Prompt Engineering: An Empirical Study
• UniGuard: Towards Universal Safety Guardrails for Jailbreak Attacks on Multimodal Large Language Models
• Survey of Vulnerabilities in Large Language Models Revealed by Adversarial Attacks
• PaliGemma 2: A Family of Versatile VLMs for Transfer
• Gemini: A Family of Highly Capable Multimodal Models
• Preventing Jailbreak Prompts as Malicious Tools for Cybercriminals: A Cyber Defense Perspective
• Cross-modality Information Check for Detecting Jailbreaking in Multimodal Large Language Models
• JailGuard: A Universal Detection Framework for LLM Prompt-based Attacks

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering