4

苹果 AI/ML 团队开发多模态大模型 Ferret,成功突破谷歌人机验证码难题

 11 months ago
source link: https://www.8btc.com/article/6835294
Go to the source link to view the article. You can view the picture content, updated content and better typesetting reading experience. If the link is broken, please click the button below to view the snapshot at that time.
neoserver,ios ssh client
2023-10-12 10:51

苹果 AI/ML 团队开发多模态大模型 Ferret,成功突破谷歌人机验证码难题

据站长之家 10 月 12 日报道,苹果 AI/ML 团队与哥伦比亚大学研究团队合作开发的多模态大模型“雪貂”(Ferret)能够在图像中准确找到交通信号灯,比 GPT-4V 表现更出色,提高了大模型在“看说答”任务中的精确度。

Ferret 的关键创新在于将引用(referring)和定位(grounding)两方面的空间理解能力紧密结合,使模型能够同时理解给定区域的语义和找到对应目标。


About Joyk


Aggregate valuable and interesting links.
Joyk means Joy of geeK