近日，上海人工智能实验室与清华大学、香港中文大学、商汤科技等机构联合发布新一代书生·视觉大模型（InternVL）。该模型的视觉编码器参数量达60亿，首次提出对比-生成融合的渐进式对齐技术，实现了在互联网级别数据上视觉大模型与语言大模型的精细对齐。此次发布的书生·视觉基础模型标志着人工智能领域的一大进步，其开源特性将促进全球研究人员和开发者在此基础上进行创新和应用。
Title: Shanghai AI Lab Unveils Cutting-Edge Visual Model
Keywords: Visual Model, Open-Source Technology, Fine Alignment
News content:
Recently, Shanghai AI Lab, in collaboration with Tsinghua University, the Chinese University of Hong Kong, and SenseTime, has released the next-generation InternVL, a visual model. The visual encoder in the new model has 6 billion parameters and introduces a progressive alignment technology that combines comparison and generation. This technology allows for fine alignment between visual models and language models on an internet scale. The release of the InternVL visual model is a significant advancement in the field of artificial intelligence, and its open-source nature will enable researchers and developers worldwide to build upon and innovate with the technology.
【来源】https://mp.weixin.qq.com/s/bdfAJRqOF9tUk8Vy9KC_XQ
Views: 1