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
☆65Updated 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☆97Updated 4 months ago
- ☆29Updated last month
- [ICML 2024] Unsupervised Adversarial Fine-Tuning of Vision Embeddings for Robust Large Vision-Language Models☆135Updated last month
- ☆102Updated last year
- [ICML 2024] Agent Smith: A Single Image Can Jailbreak One Million Multimodal LLM Agents Exponentially Fast☆107Updated last year
- [ArXiv 2024] Denial-of-Service Poisoning Attacks on Large Language Models☆19Updated 8 months ago
- [ECCV'24 Oral] The official GitHub page for ''Images are Achilles' Heel of Alignment: Exploiting Visual Vulnerabilities for Jailbreaking …☆25Updated 8 months ago
- ☆52Updated 3 months ago
- [COLM 2024] JailBreakV-28K: A comprehensive benchmark designed to evaluate the transferability of LLM jailbreak attacks to MLLMs, and fur…☆68Updated 2 months ago
- A toolbox for benchmarking trustworthiness of multimodal large language models (MultiTrust, NeurIPS 2024 Track Datasets and Benchmarks)☆154Updated 3 weeks ago
- [AAAI'25 (Oral)] Jailbreaking Large Vision-language Models via Typographic Visual Prompts☆156Updated 3 weeks ago
- ☆34Updated 3 months ago
- Official codebase for "STAIR: Improving Safety Alignment with Introspective Reasoning"☆57Updated 4 months ago
- [ICLR 2024 Spotlight 🔥 ] - [ Best Paper Award SoCal NLP 2023 🏆] - Jailbreak in pieces: Compositional Adversarial Attacks on Multi-Modal…☆61Updated last year
- Code for Neurips 2024 paper "Shadowcast: Stealthy Data Poisoning Attacks Against Vision-Language Models"☆50Updated 6 months ago
- [ICLR 2024] Inducing High Energy-Latency of Large Vision-Language Models with Verbose Images☆36Updated 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
- To Think or Not to Think: Exploring the Unthinking Vulnerability in Large Reasoning Models☆31Updated last month
- [ECCV 2024] The official code for "AdaShield: Safeguarding Multimodal Large Language Models from Structure-based Attack via Adaptive Shi…☆60Updated last year
- A package that achieves 95%+ transfer attack success rate against GPT-4☆23Updated 8 months ago
- [ECCV'24 Oral] The official GitHub page for ''Images are Achilles' Heel of Alignment: Exploiting Visual Vulnerabilities for Jailbreaking …☆29Updated 9 months ago
- ☆47Updated last year
- [ACL 2025] Data and Code for Paper VLSBench: Unveiling Visual Leakage in Multimodal Safety☆48Updated 2 months ago
- [ICML 2024] Safety Fine-Tuning at (Almost) No Cost: A Baseline for Vision Large Language Models.☆74Updated 5 months ago
- Code to conduct an embedding attack on LLMs☆25Updated 6 months ago
- Panda Guard is designed for researching jailbreak attacks, defenses, and evaluation algorithms for large language models (LLMs).☆36Updated last month
- Code for "When LLM Meets DRL: Advancing Jailbreaking Efficiency via DRL-guided Search" (NeurIPS 2024)☆9Updated 8 months ago
- ☆93Updated 5 months ago
- ☆22Updated last year
- ☆30Updated 2 months ago