上海人工智能实验室（上海AI实验室）近日联合清华大学、香港中文大学、商汤科技等机构，发布新一代“书生·视觉大模型”（InternVL）。这一模型的视觉编码器参数量达到60亿（InternVL-6B），并在互联网级别数据上，首次提出并实现了对比-生成融合的渐进式对齐技术，实现了视觉大模型与语言大模型的精细对齐。
InternVL的发布，标志着我国在人工智能领域取得了重要突破。开源领先的技术，不仅为我国人工智能研究提供了强大的技术支持，也为全球人工智能发展贡献了中国智慧。
英文标题Title：Shanghai Releases Open-source Leading Visual Large Model
英文关键词Keywords：AI Lab, Visual Large Model, Open-source Leading
英文新闻内容News content：The Shanghai Artificial Intelligence Laboratory (Shanghai AI Lab) recently joined forces with Tsinghua University, the Chinese University of Hong Kong, SenseTime, and other institutions to release the next-generation “Scholar Visual Large Model” (InternVL). The visual encoder of this model has a parameter quantity of 6 billion (InternVL-6B), and for the first time, it proposes and implements a comparative-generative fusion progressive alignment technology on internet-scale data, achieving fine alignment between visual large models and language large models.
The release of InternVL marks an important breakthrough in the field of artificial intelligence in China. The open-source leading technology not only provides strong technical support for artificial intelligence research in our country but also contributes Chinese wisdom to the global development of artificial intelligence.
【来源】https://mp.weixin.qq.com/s/bdfAJRqOF9tUk8Vy9KC_XQ
Views: 1