Fudan University
A well-known comprehensive research university in China, with significant research contributions in fields such as artificial intelligence and computer science.
Timeline (last 90 days)
Fudan University, together with the OpenMouse team, open-sourced the 5B image generation model Mouse Image, which natively supports ultra-long prompts of 4000 characters, can precisely render dense Chinese and English characters, and supports structured code control of image layout.
Jia Xiaosong is an assistant professor at the Institute of Trustworthy Embodied Intelligence, Fudan University.
Sun, Xu, Shi, and Yang of Fudan University and the University of Science and Technology of China tested three models, Jev, Laya, and Open-Jev, in the paper "Type-Safe Is Not Error-Free" (September 2026).
Teams from Fudan University, Tencent Hunyuan, and Zhejiang University open-sourced the audio-video joint generation framework Prism, which can natively generate videos with sound at resolutions from 720P and 1080P to 2K.
Tencent Hunyuan, in collaboration with Fudan University and Zhejiang University, open-sourced Prism under the MIT license; by reducing redundant computation in high-resolution training, it achieves approximately 2.5× speedup in joint native 2K video and audio generation training.
Tencent Membership, in collaboration with Fudan University and Tsinghua University, released the exploration capability evaluation tool Exploration Bench, which constructs an alien world that conflicts with common sense and has verifiable rules, requiring AI to correct its hypotheses through four rounds of experiments.
复旦大学自然语言处理团队于2023年发表综述论文《A Survey on Large Language Model based Autonomous Agents》
All Facts (7)
Jia Xiaosong is an assistant professor at the Institute of Trustworthy Embodied Intelligence, Fudan University.
50%Unverified复旦大学自然语言处理团队于2023年发表综述论文《A Survey on Large Language Model based Autonomous Agents》
50%UnverifiedFudan University, together with the OpenMouse team, open-sourced the 5B image generation model Mouse Image, which natively supports ultra-long prompts of 4000 characters, can precisely render dense Chinese and English characters, and supports structured code control of image layout.
50%UnverifiedSun, Xu, Shi, and Yang of Fudan University and the University of Science and Technology of China tested three models, Jev, Laya, and Open-Jev, in the paper "Type-Safe Is Not Error-Free" (September 2026).
50%UnverifiedTeams from Fudan University, Tencent Hunyuan, and Zhejiang University open-sourced the audio-video joint generation framework Prism, which can natively generate videos with sound at resolutions from 720P and 1080P to 2K.
50%UnverifiedTencent Hunyuan, in collaboration with Fudan University and Zhejiang University, open-sourced Prism under the MIT license; by reducing redundant computation in high-resolution training, it achieves approximately 2.5× speedup in joint native 2K video and audio generation training.
50%UnverifiedTencent Membership, in collaboration with Fudan University and Tsinghua University, released the exploration capability evaluation tool Exploration Bench, which constructs an alien world that conflicts with common sense and has verifiable rules, requiring AI to correct its hypotheses through four rounds of experiments.
50%