VILA-Lab / M-AttackLinks
A Simple Baseline Achieving Over 90% Success Rate Against the Strong Black-box Models of GPT-4.5/4o/o1. Paper at: https://arxiv.org/abs/2503.10635
☆66Updated 3 months ago
Alternatives and similar repositories for M-Attack
Users that are interested in M-Attack are comparing it to the libraries listed below
Sorting:
- [ICLR 2025] Dissecting adversarial robustness of multimodal language model agents☆98Updated 5 months ago
- ☆32Updated 2 months ago
- [COLM 2024] JailBreakV-28K: A comprehensive benchmark designed to evaluate the transferability of LLM jailbreak attacks to MLLMs, and fur…☆75Updated 3 months ago
- [AAAI'25 (Oral)] Jailbreaking Large Vision-language Models via Typographic Visual Prompts☆160Updated last month
- [ICLR 2024 Spotlight 🔥 ] - [ Best Paper Award SoCal NLP 2023 🏆] - Jailbreak in pieces: Compositional Adversarial Attacks on Multi-Modal…☆65Updated last year
- [ECCV 2024] Official PyTorch Implementation of "How Many Unicorns Are in This Image? A Safety Evaluation Benchmark for Vision LLMs"☆81Updated last year
- Code for Neurips 2024 paper "Shadowcast: Stealthy Data Poisoning Attacks Against Vision-Language Models"☆51Updated 6 months ago
- A toolbox for benchmarking trustworthiness of multimodal large language models (MultiTrust, NeurIPS 2024 Track Datasets and Benchmarks)☆158Updated last month
- A package that achieves 95%+ transfer attack success rate against GPT-4☆23Updated 9 months ago
- ☆37Updated 4 months ago
- [ICML 2024] Unsupervised Adversarial Fine-Tuning of Vision Embeddings for Robust Large Vision-Language Models☆138Updated 2 months ago
- [ICLR 2024] Inducing High Energy-Latency of Large Vision-Language Models with Verbose Images☆37Updated last year
- [ECCV 2024] The official code for "AdaShield: Safeguarding Multimodal Large Language Models from Structure-based Attack via Adaptive Shi…☆62Updated last year
- ☆102Updated last year
- [ICML 2024] Safety Fine-Tuning at (Almost) No Cost: A Baseline for Vision Large Language Models.☆73Updated 6 months ago
- ☆55Updated 4 months ago
- ☆49Updated last year
- ☆31Updated 2 months ago
- [CVPR2025] Official Repository for IMMUNE: Improving Safety Against Jailbreaks in Multi-modal LLMs via Inference-Time Alignment☆22Updated last month
- A Survey on Jailbreak Attacks and Defenses against Multimodal Generative Models☆215Updated last month
- To Think or Not to Think: Exploring the Unthinking Vulnerability in Large Reasoning Models☆31Updated 2 months ago
- [ECCV'24 Oral] The official GitHub page for ''Images are Achilles' Heel of Alignment: Exploiting Visual Vulnerabilities for Jailbreaking …☆34Updated 9 months ago
- Panda Guard is designed for researching jailbreak attacks, defenses, and evaluation algorithms for large language models (LLMs).☆42Updated last week
- Official codebase for Image Hijacks: Adversarial Images can Control Generative Models at Runtime☆50Updated last year
- [ArXiv 2024] Denial-of-Service Poisoning Attacks on Large Language Models☆20Updated 9 months ago
- Accepted by ECCV 2024☆147Updated 9 months ago
- ☆46Updated 8 months ago
- Code to conduct an embedding attack on LLMs☆27Updated 6 months ago
- [ECCV'24 Oral] The official GitHub page for ''Images are Achilles' Heel of Alignment: Exploiting Visual Vulnerabilities for Jailbreaking …☆26Updated 9 months ago
- ☆28Updated 4 months ago