VILA-Lab / M-Attack
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
☆56Updated 3 weeks ago
Alternatives and similar repositories for M-Attack
Users that are interested in M-Attack are comparing it to the libraries listed below
Sorting:
- [ICML 2024] Unsupervised Adversarial Fine-Tuning of Vision Embeddings for Robust Large Vision-Language Models☆130Updated 6 months ago
- [ICLR 2025] Dissecting adversarial robustness of multimodal language model agents☆84Updated 2 months ago
- ☆32Updated last month
- [ICML 2024] Safety Fine-Tuning at (Almost) No Cost: A Baseline for Vision Large Language Models.☆71Updated 3 months ago
- A package that achieves 95%+ transfer attack success rate against GPT-4☆19Updated 6 months ago
- [ECCV 2024] Official PyTorch Implementation of "How Many Unicorns Are in This Image? A Safety Evaluation Benchmark for Vision LLMs"☆80Updated last year
- [ECCV 2024] The official code for "AdaShield: Safeguarding Multimodal Large Language Models from Structure-based Attack via Adaptive Shi…☆57Updated 10 months ago
- [ECCV'24 Oral] The official GitHub page for ''Images are Achilles' Heel of Alignment: Exploiting Visual Vulnerabilities for Jailbreaking …☆29Updated 6 months ago
- ☆96Updated last year
- A toolbox for benchmarking trustworthiness of multimodal large language models (MultiTrust, NeurIPS 2024 Track Datasets and Benchmarks)☆145Updated last month
- [ICML 2024] Agent Smith: A Single Image Can Jailbreak One Million Multimodal LLM Agents Exponentially Fast☆101Updated last year
- ☆21Updated last month
- Official codebase for "STAIR: Improving Safety Alignment with Introspective Reasoning"☆36Updated 2 months ago
- [ECCV'24 Oral] The official GitHub page for ''Images are Achilles' Heel of Alignment: Exploiting Visual Vulnerabilities for Jailbreaking …☆19Updated 6 months ago
- ECSO (Make MLLM safe without neither training nor any external models!) (https://arxiv.org/abs/2403.09572)☆23Updated 6 months ago
- ☆41Updated last month
- [AAAI'25 (Oral)] Jailbreaking Large Vision-language Models via Typographic Visual Prompts☆136Updated 2 months ago
- ☆27Updated last year
- [ICLR 2024 Spotlight 🔥 ] - [ Best Paper Award SoCal NLP 2023 🏆] - Jailbreak in pieces: Compositional Adversarial Attacks on Multi-Modal…☆52Updated 11 months ago
- [ICML 2024 Oral] Official code repository for MLLM-as-a-Judge.☆67Updated 2 months ago
- Accepted by ECCV 2024☆129Updated 6 months ago
- ☆86Updated last year
- Code for Neurips 2024 paper "Shadowcast: Stealthy Data Poisoning Attacks Against Vision-Language Models"☆46Updated 3 months ago
- [ICLR 2024] Inducing High Energy-Latency of Large Vision-Language Models with Verbose Images☆33Updated last year
- [ArXiv 2024] Denial-of-Service Poisoning Attacks on Large Language Models☆18Updated 6 months ago
- [COLM 2024] JailBreakV-28K: A comprehensive benchmark designed to evaluate the transferability of LLM jailbreak attacks to MLLMs, and fur…☆60Updated this week
- The official code of the paper "Deciphering Cross-Modal Alignment in Large Vision-Language Models with Modality Integration Rate".☆98Updated 5 months ago
- [NeurIPS-2023] Annual Conference on Neural Information Processing Systems☆192Updated 4 months ago
- Improved Few-Shot Jailbreaking Can Circumvent Aligned Language Models and Their Defenses (NeurIPS 2024)☆61Updated 4 months ago
- ☆40Updated 11 months ago