我国上海人工智能实验室与商汤科技联合香港中文大学和复旦大学，近日正式发布新一代大语言模型书生·浦语2.0。这一版本的核心理念在于回归语言建模的本质，旨在通过提升语料质量及信息密度，使模型基座的语言建模能力得到质的飞跃。书生·浦语2.0具备强大的处理能力，支持200K token的上下文，可以一次性接收并处理约30万汉字的输入内容，准确提取关键信息，实现长文本中的“大海捞针”。
书生·浦语2.0的发布，标志着我国在人工智能领域的研究取得了新的突破。这款大语言模型不仅为各类语言任务提供了强大的支持，还为实现人工智能与人类语言的深度融合奠定了基础。作为一款开放源代码的软件，书生·浦语2.0将为广大研究者提供一个宝贵的实践平台，推动我国人工智能事业的发展。
Title: Shanghai AI Lab Releases Sheng·Puyu 2.0, Returning to the Essence of Language Modeling
Keywords: Artificial Intelligence, Large Language Model, Sheng·Puyu 2.0
News Content:
Shanghai Artificial Intelligence Laboratory, together with SenseTime, jointly released the new generation of large language model Sheng·Puyu 2.0 with Hong Kong Chinese University and Fudan University. The core concept of Sheng·Puyu 2.0 lies in returning to the essence of language modeling, aiming to improve the quality of corpus and information density, thus achieving a qualitative leap in the language modeling capabilities of the model base. Sheng·Puyu 2.0 has strong processing capabilities, supporting 200K token context, and can receive and process about 300,000 Chinese characters input content at once, accurately extracting key information and realizing “fishing in a sea of needles” in long texts.
The release of Sheng·Puyu 2.0 marks a new breakthrough in the field of artificial intelligence in our country. This large language model not only provides strong support for various language tasks but also lays the foundation for the deep integration of artificial intelligence and human language. As an open-source software, Sheng·Puyu 2.0 will provide a valuable practice platform for researchers, promoting the development of artificial intelligence in our country.
【来源】https://www.thepaper.cn/newsDetail_forward_26040295
Views: 1