Atlas / Reports / Detail
Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling
Multimodal Large Language Models report from DeepSeek with 13 connected researchers in the LLMpeople atlas.
Connected researchers
Researcher
3 reports
Jifeng Dai
DeepSeek / MiniMax
Researcher focused on computer vision, multimodal learning, and generative AI. His public homepage says he is currently with Stepfun, after serving as a principal scientist at SenseTime Research and a researcher at Microsoft Research Asia, and that he earned a PhD in computer science from Tsinghua University.
Researcher
6 reports
Huazuo Gao
DeepSeek
Researcher at DeepSeek AI working on decision-making and post-training for large language models.
Researcher
1 reports
Xiangkun Wang
DeepSeek
Research intern at DeepSeek and undergraduate student at Tsinghua University focusing on multimodal large language models, agents, and embodied AI.
Researcher
1 reports
Zezhou Wang
DeepSeek
Research intern at DeepSeek and master's student at Tsinghua University working on large language models, reinforcement learning, and multimodal understanding and generation.
Researcher
1 reports
Jinghong Yuan
DeepSeek
PhD student at UC San Diego researching reasoning, planning, and multimodal foundation models; publication context connects Jinghong Yuan to Janus-Pro.
Researcher
1 reports
Jiaxuan Fan
DeepSeek
Jiaxuan Fan is a machine learning researcher at DeepSeek. Her interests include data-centric AI, model efficiency, and multimodal learning.
Researcher
5 reports
Binyuan Hui
DeepSeek / Alibaba Qwen
AI researcher whose public work includes large language models, vision-language models, and multimodal systems. His public profile notes prior work as a senior algorithm expert at Alibaba and co-authorship of Qwen technical reports.
Researcher
1 reports
Xiaoze Liu
DeepSeek
Research intern at DeepSeek and PhD student at Carnegie Mellon University interested in machine learning, agents, language, vision, robotics, and healthcare.
Researcher
1 reports
Xinyu Li
DeepSeek
Research intern at DeepSeek and undergraduate student at Tsinghua University working on vision-language models, inference-time scaling, and reinforcement learning.
Researcher
1 reports
Zhihuan Liu
DeepSeek
Research intern at DeepSeek and PhD student at Shanghai Jiao Tong University working on large language models, reasoning, agents, and reinforcement learning.
Researcher
1 reports
Hongxia Yang
DeepSeek
External advisor at DeepSeek and former Corporate Vice President and Chief Scientist at Microsoft Research Asia.
Researcher
5 reports
Jie Zhou
DeepSeek / Moonshot AI
Jie Zhou is a Moonshot AI contributor and co-author of Kimi k1.5: Scaling Reinforcement Learning with LLMs.
Researcher
2 reports
Shang Yang
DeepSeek / MiniMax
Researcher focused on reinforcement learning, large language model reasoning, and multimodal foundation models; coauthor of Janus-Pro and MiniMax-M1.