在近日举行的2024环球时报年会“创新、科技与人文：中国式现代化的未来图景”议题讨论中，著名人工智能和大数据专家、清华大学电子工程系兼职教授窦德景表示，我国国内大模型目前已达到GPT-3.5水平。虽然与国际领先技术GPT-4还有一定差距，但这一差距正在逐步缩小。这预示着我国在人工智能领域的研究正不断迈向更高水平。
窦德景教授的观点凸显出我国在人工智能技术研发上的快速进步。大模型是人工智能领域的重要研究方向，GPT-3.5和GPT-4代表了当前全球最高技术水平。我国能在短时间内达到GPT-3.5水平，显示出国内科研团队在人工智能技术研发上的突破和创新。
随着技术差距的缩小，我国在人工智能领域的国际竞争力不断提升。未来，我国将继续深化人工智能技术研究，推动科技与人文的融合，探索中国式现代化的未来图景。
英文翻译：
News Title: Domestic large models approach GPT-3.5, gap with GPT-4 shrinking
Keywords: Artificial intelligence, large models, technical gap
News Content:
In the recent 2024 Global Times Annual Meeting discussion on the topic “Innovation, Technology, and Humanity: The Future Landscape of Chinese Modernization”, famous artificial intelligence and big data expert Professor Dou Dejing, a part-time professor at the Department of Electronic Engineering, Tsinghua University, said that domestic large models have reached the GPT-3.5 level. Although there is still a certain gap with the international leading technology GPT-4, the gap is gradually narrowing. This indicates that China’s research in artificial intelligence is continuously moving towards a higher level.
Professor Dou’s opinion highlights the rapid progress of artificial intelligence technology research and development in China. Large models are an important research direction in the field of artificial intelligence, and GPT-3.5 and GPT-4 represent the highest technical levels globally. Reaching the GPT-3.5 level in a short period of time demonstrates the breakthroughs and innovations of domestic research teams in artificial intelligence technology research.
As the technical gap narrows, China’s international competitiveness in the field of artificial intelligence continues to rise. In the future, China will continue to deepen research on artificial intelligence technology, promote the integration of technology and humanity, and explore the future landscape of Chinese modernization.
【来源】https://china.huanqiu.com/article/4FsziDwnvnj
Views: 1