目录 #
Stage1: 模块独立[2] #
{% asset_img ’’ %}
model #
- CLIP
- ViLT
- ALBEF
Stage2: 模块共享[2] #
model #
- VLMO
- BLIP
- BLIP2
- BEiTv3
Stage3: 范式统一[2] #
model #
- Unified-IO
- Uni-Perceiver
- PaLi
总结 [1] #
{% asset_img ’’ %}
参考 #
Overview #
1xx. 多模态论文串讲 *** 多模态论文串讲：ALBEF & VLMo & BLIP & CoCa & Beit V3
1xx. 图生文多模态大模型开源项目回顾：兼看20240307大模型进展早报
1xx. 图文多模态大模型综述
1xx. Multimodality and Large Multimodal Models (LMMs) 多模态和多模态大模型 (LMM)[译] CLIP Flamingo