ViLT

2000년 도입 · 논문 6편에서 사용

ViLT is a minimal vision-and-language pre-training transformer model where processing of visual inputs is simplified to just the same convolution-free manner that text inputs are processed. The model-specific components of ViLT require less computation than the transformer component for multimodal interactions. ViLTThe model is pre-trained on the following objectives: image text matching, masked language modeling, and word patch alignment.

출처: ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision

소개 논문: ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision