erfanshayegani / Jailbreak-In-Pieces
[ICLR 2024 Spotlight 🔥 ] - [ Best Paper Award SoCal NLP 2023 🏆] - Jailbreak in pieces: Compositional Adversarial Attacks on Multi-Modal Language Models
☆14Updated 3 months ago
Related projects: ⓘ
- ☆30Updated last month
- Code for the paper "Shadowcast: Stealthy Data Poisoning Attacks Against Vision-Language Models"☆21Updated 5 months ago
- [ICLR 2024] Inducing High Energy-Latency of Large Vision-Language Models with Verbose Images☆17Updated 7 months ago
- ☆13Updated 3 months ago
- Accepted by ECCV 2024☆59Updated 2 months ago
- [ECCV 2024] The official code for "AdaShield: Safeguarding Multimodal Large Language Models from Structure-based Attack via Adaptive Shi…☆34Updated 2 months ago
- ☆22Updated 2 months ago
- ☆24Updated 3 months ago
- ☆14Updated 2 months ago
- JailBreakV-28K: A comprehensive benchmark designed to evaluate the transferability of LLM jailbreak attacks to MLLMs, and further assess …☆29Updated 2 months ago
- This is an official repository of ``VLAttack: Multimodal Adversarial Attacks on Vision-Language Tasks via Pre-trained Models'' (NeurIPS 2…☆29Updated 8 months ago
- One Prompt Word is Enough to Boost Adversarial Robustness for Pre-trained Vision-Language Models☆31Updated 4 months ago
- [ECCV 2024] Official PyTorch Implementation of "How Many Unicorns Are in This Image? A Safety Evaluation Benchmark for Vision LLMs"☆61Updated 9 months ago
- [CVPR23W] "A Pilot Study of Query-Free Adversarial Attack against Stable Diffusion" by Haomin Zhuang, Yihua Zhang and Sijia Liu☆21Updated 3 weeks ago
- ☆35Updated last year
- The official implementation of ECCV'24 paper "To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Uns…☆43Updated last month
- [NeurIPS-2023] Annual Conference on Neural Information Processing Systems☆147Updated 10 months ago
- [ECCV-2024] Transferable Targeted Adversarial Attack, CLIP models, Generative adversarial network, Multi-target attacks☆18Updated last month
- [ICML 2024] Safety Fine-Tuning at (Almost) No Cost: A Baseline for Vision Large Language Models.☆36Updated 3 weeks ago
- AnyDoor: Test-Time Backdoor Attacks on Multimodal Large Language Models☆39Updated 5 months ago
- ☆84Updated 7 months ago
- ☆14Updated 4 months ago
- Improved techniques for optimization-based jailbreaking on large language models☆33Updated 3 months ago
- [CVPR'24 Oral] Metacloak: Preventing Unauthorized Subject-driven Text-to-image Diffusion-based Synthesis via Meta-learning☆18Updated 2 months ago
- Implementation of BadCLIP https://arxiv.org/pdf/2311.16194.pdf☆14Updated 5 months ago
- [MM'23 Oral] "Text-to-image diffusion models can be easily backdoored through multimodal data poisoning"☆19Updated last week
- Set-level Guidance Attack: Boosting Adversarial Transferability of Vision-Language Pre-training Models. [ICCV 2023 Oral]☆45Updated last year
- Official PyTorch implementation of "CleanCLIP: Mitigating Data Poisoning Attacks in Multimodal Contrastive Learning" @ ICCV 2023☆27Updated 8 months ago
- ☆18Updated 3 months ago
- [CVPR 2023] Backdoor Defense via Adaptively Splitting Poisoned Dataset☆40Updated 5 months ago