Files
2026-08-30 15:36:28 +08:00

14 lines
929 B
Markdown
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
# Flamingo: a Visual Language Model for Few-Shot Learning
- 第一作者:Jean-Baptiste Alayrac
- 发表:NeurIPS 2022 主会
- 阅读深度:摘要及 gated cross-attention 方法片段筛选
- 记录日期:2026-08-28
- 来源:[正式条目](https://proceedings.neurips.cc/paper_files/paper/2022/hash/960a172bc7fbf0177ccccbb411a7d800-Abstract-Conference.html)、[论文](https://arxiv.org/abs/2204.14198)
## 摘要与项目关系
利用桥接视觉和语言模型的结构,使模型能处理交错图像/视频及文本;通过新增视觉交互模块利用预训练骨干。它是门控 cross-attention 内部融合的基础参考,不是本项目需要从头重训的完整系统。
本项目拟借鉴有限视觉 memory 与局部 adapter,不声称新增 gate 本身构成方法创新。更直接的 AV 迁移证据参见 Whisper-Flamingo 笔记;接话因果性和人物交互仍需本项目独立证明。