yunqing-me / AttackVLM

[NeurIPS-2023] Annual Conference on Neural Information Processing Systems

https://arxiv.org/pdf/2305.16934.pdf

yunqing-me/AttackVLM Issues

Segmentation fault when doing `clip_model.encode_image`
Updated 2 months ago
It seems that it can only be run on a single card. Is there a multi-card version?
Updated 2 months ago
Is your method black-box or gray-box？
Closed 4 months ago1
What models are used for img2prompt and LLAVA
Updated 4 months ago
Questions about attack on BLIP (LAVIS)
Closed 9 months ago3
Blip2OPT' object has no attribute 'forward_encoder_image'
Updated 5 months ago
Reproducing MF-it result in Table 2 of the paper
Closed 8 months ago4
Where are the implementations of MF-it?
Updated 7 months ago
可以用3090嘛
Closed 7 months ago1
Why only using the first dimension of the image encoding?
Closed 7 months ago
dimension mismatch in code
Closed 7 months ago3
Question regarding the generation of adversarial examples
Closed 7 months ago1
Regarding data path in LAVIS
Closed 8 months ago1
The large model you just updated includes Unidiff, MiniGPT-4 and LAVIS, but it is not mentioned in the LAVIS paper. I would like to ask what model this is, and how to reproduce several large models in the paper?For example, BLIP, BLIP2, etc.
Closed 9 months ago1
MiniGPT-4 Code
Closed 9 months ago1
Connot interpret 'torch.float32' as a data type
Closed 9 months ago1
VLM model
Closed 10 months ago2
data question
Closed a year ago5
Import package problem
Closed a year ago
We couldn't connect to 'https://huggingface.co' to load this model, couldn't find it in the cached files and it looks like CompVis/stable-diffusion-safety-checker is not the path to a directory containing a config.json file. Checkout your internet connection or see how to run the library in offline mode at 'https://huggingface.co/docs/transformers/installation#offline-mode'
Closed a year ago6