**生数科技携手清华大学发布国内首个长时长视频大模型「Vidu」**
今日，在中关村论坛未来人工智能先锋论坛上，生数科技与清华大学联合发布了一项具有重大创新意义的科研成果——中国首个长时长视频大模型「Vidu」。该模型的发布标志着我国在人工智能领域，特别是在视频生成技术上取得了显著的进步。
据悉，此款视频大模型由生数科技与清华大学团队共同研发，采用了团队原创的Diffusion与Transformer融合的架构U-ViT。U-ViT架构的独到之处在于，它不仅继承了Diffusion模型在图像生成上的优势，还融合了Transformer模型在处理视频序列上的特点，使得模型在生成长时长、高一致性、高动态性的视频内容上表现出色。
更值得一提的是，「Vidu」模型具有强大的一键生成能力。用户只需简单操作，即可生成长达16秒、分辨率高达1080P的高清视频内容。这一功能为短视频制作、实时直播、教育培训等多个领域提供了前所未有的便利。
此次「Vidu」模型的发布，不仅体现了我国在人工智能技术上的创新与突破，也为各行各业带来了全新的应用场景和想象空间。未来，生数科技与清华大学将继续深化合作，共同推动我国人工智能技术的发展，为人类社会带来更多价值。
在中关村论坛未来人工智能先锋论坛上，专家们对「Vidu」模型给予了高度评价，认为它代表了我国在人工智能领域的一个重要里程碑，有望引领全球视频生成技术的发展趋势。
英语如下：
Certainly, here is the translation in English using Markdown format:
“`markdown
### News Title: ShenShu Technology Collaborates with Tsinghua University to Launch “Vidu” Long Video Large Model
Keywords: ShenShu Technology, Tsinghua University, Launches Vidu Model.
#### ShenShu Technology and Tsinghua University Unveil China’s First Long-Duration Video Large Model ‘Vidu’
Today, at the Future Artificial Intelligence Pioneer Forum of the Zhongguancun Forum, ShenShu Technology and Tsinghua University jointly released a significant scientific research achievement — China’s first long-duration video large model ‘Vidu.’ The launch of this model signifies significant progress made by China in the field of artificial intelligence, particularly in video generation technology.
It is understood that this video large model was co-developed by the teams at ShenShu Technology and Tsinghua University, utilizing their original architecture U-ViT, which integrates Diffusion and Transformer. The uniqueness of the U-ViT architecture lies in its ability to inherit the advantages of the Diffusion model in image generation while integrating the characteristics of the Transformer model in processing video sequences, resulting in exceptional performance in generating long-duration, high-consistency, and high-dynamic video content.
What’s more notable is the ‘Vidu’ model’s powerful one-click generation capability. Users can easily generate high-definition video content up to 16 seconds long and with a resolution of 1080P. This feature offers unprecedented convenience for short video production, live streaming, educational training, and other fields.
The launch of the ‘Vidu’ model not only reflects China’s innovation and breakthroughs in artificial intelligence technology but also brings new application scenarios and possibilities to various industries. In the future, ShenShu Technology and Tsinghua University will continue to deepen their collaboration, jointly promoting the development of China’s artificial intelligence technology and bringing more value to human society.
At the Future Artificial Intelligence Pioneer Forum of the Zhongguancun Forum, experts highly praised the ‘Vidu’ model, considering it an important milestone of China’s AI field and expected to lead the global trend in video generation technology.
“`
This translation maintains the structured format of the original text and conveys the technical and collaborative achievements of the Chinese entities in the realm of artificial intelligence.
【来源】https://mp.weixin.qq.com/s/Lba_WJTQWjQ3lOaLNJA4qA
Views: 1