近日，我国上海人工智能实验室（Shanghai AI Laboratory）联合清华大学、香港中文大学、商汤科技等机构，成功开源了新一代书生·视觉大模型（InternVL）。这一举措在我国人工智能领域取得了重要突破。
新一代书生·视觉基础模型（InternVL-6B）的视觉编码器参数量高达60亿，首次提出了对比-生成融合的渐进式对齐技术。该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，进一步推动了人工智能的发展。
此次开源的新一代书生·视觉大模型，不仅在技术上领先，更是对全球人工智能研究的重要贡献。这一成果的发布，展示了我国在人工智能领域的强大实力，也为全球研究者提供了一个全新的研究平台。
英文翻译：
News Title: Next-Generation Scholar Visual Large Model Achieves Leadership in Open Source
Keywords: Shanghai AI Laboratory, Scholar Visual Large Model, Open Source
News Content:
Recently, the Shanghai AI Laboratory, in collaboration with Tsinghua University, the Chinese University of Hong Kong, and SenseTime, has successfully opened up the next-generation Scholar Visual Large Model (InternVL). This breakthrough marks an important milestone in China’s artificial intelligence (AI) field.
The next-generation Scholar Visual Basic Model (InternVL-6B) features a visual encoder with an impressive 6 billion parameters, pioneering the contrast-generation fusion progressive alignment technology. This technology achieves fine alignment between visual and language models at internet scale, further propelling the development of AI.
The open-sourcing of the next-generation Scholar Visual Large Model not only represents a technological lead but also contributes significantly to global AI research. The release of this achievement demonstrates China’s strong presence in the AI field and provides a new research platform for global researchers.
【来源】https://mp.weixin.qq.com/s/bdfAJRqOF9tUk8Vy9KC_XQ
Views: 1