Year: 2026
Type: preprint
Abstract: We introduce Yuan3.0 Flash, an open-source Mixture-of-Experts (MoE) MultiModal Large Language Model featuring 3.7B activated parameters and 40B total parameters, specifically designed to enhance performance on enterprise-oriented tasks while maintaining competitive capabilities on general-purpose tasks. To address the overthinking phenomenon commonly observed in Large Reasoning Models (LRMs), we propose Reflection-aware Adaptive Policy Optimization (RAPO), a novel RL training algorithm that effectively regulates overthinking behaviors. In enterprise-oriented tasks such as retrieval-augmented generation (RAG), complex table understanding, and summarization, Yuan3.0 Flash consistently achieves superior performance. Moreover, it also demonstrates strong reasoning capabilities in domains such as mathematics, science, etc., attaining accuracy comparable to frontier model while requiring only approximately 1/4 to 1/2 of the average tokens. Yuan3.0 Flash has been fully open-sourced to facilitate further research and real-world deployment: https://github.com/Yuan-lab-LLM/Yuan3.0.
Source: arXiv (Cornell University)
Authors: ai, YuanLab., :, Wu, Shawn, Wang, Sean, Li, Louie, Chen, Darcy, Wang, Allen, Luo, Jiangang, Zhao, Xudong, Shen, Joseph, Ma, Gawain, Jia, Jasper, Mao, Marcus, Wang, Claire, He, Hunter, Wang, Carol, Zhang, Zera, Wang, Jason, Shen, Chonly, Zhang, Leo, Chen, Logan, Meng, Qasim, Gong, James, Zhao, Danied, Zheng, Penn, Zhu, Owen, Yu, Tong
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Funder: University of Pennsylvania
OpenAlex ID: W7119007231