PrototypeCommon
VILA-Lab/M-Attack
[NeurIPS25 & ICML25 Workshop on Reliable and Responsible Foundation Models] 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
- Owner
- VILA-Lab (organization)
- Stars
- Stars: 103
- Forks
- Forks: 6
- Language
- Python
- License
- MIT
- Created
- 2025
- Topics
- attacklvlmsmllmsadversarial-attack
- Tier
- Prototype (100 to 249 ★) What this tier means
More repos by VILA-Lab
All 6 repos by VILA-Lab- VILA-Lab/Dive-into-Claude-CodeStars: 2,206RC
- VILA-Lab/Awesome-DLMsStars: 1,244Beta
- VILA-Lab/ATLASStars: 991PythonBeta
- VILA-Lab/FigMirrorStars: 558PythonAlpha
- VILA-Lab/SRe2LStars: 142PythonPrototype
Nearby in Python
All Python repos- Mcsd7/ReaDroidStars: 103PythonPrototype
- code-zm/plex2shellStars: 103PythonPrototype
- akitaonrails/worldreportsStars: 103PythonPrototype
- mdirolf/DjanMonStars: 103PythonPrototype
- shrubberysoft/django-picklefieldStars: 103PythonPrototype
- DanielStutzbach/heapdictStars: 103PythonPrototype