Moderating Illicit Online Image Promotion for Unsafe User Generated Content Games Using Large Vision-Language Models.
Keyan GuoAyush UtkarshWenbo DingIsabelle OndracekZiming ZhaoGuo FreemanNishant VishwamitraHongxin HuPublished in: USENIX Security Symposium (2024)
Keyphrases
- language model
- user generated content
- language modeling
- image classification
- image features
- image representation
- image content
- online forums
- social media
- probabilistic model
- n gram
- speech recognition
- image collections
- document retrieval
- image regions
- language modelling
- statistical language models
- information retrieval
- image retrieval
- smoothing methods
- user generated
- query expansion
- test collection