Shuo Gao

I am a Computer Science M.S. student at Renmin University of China. My research focuses on multimodal learning and Vision-Language Models. I am also interested in robotics and embodied intelligence.

I am currently a research intern at LinkerBot, working on Embodied Foundation Models and World Action Models for Dexterous Manipulations. I also work at RAIR Lab, UNC-Chapel Hill as a remote intern, advised by Professor Zhongzheng (Jason) Ren. Previously, I have spent time at 360 AI Research, the AIM Lab at the University of Amsterdam, and the AIMC Lab at Renmin University of China.

profile image
News
  • 09/2026 🎉 Our paper Elephant in the Fridge: Constant-Memory Frame Packing for Long Video Understanding has been accepted to NeurIPS 2026!
  • 07/2026 Started as a research intern at LinkerBot, working on Embodied Foundation Models.
  • 02/2026 Started as a remote intern at RAIR Lab, UNC-Chapel Hill, working on long video understanding VLMs.
  • 10/2025 Started a research internship at 360 AI Research, working on MLLM-based precise object detection.
  • 01/2025 Started as a remote intern at AIM Lab, University of Amsterdam, working on uncertainty-guided test-time prompt adaptation for VLMs.
  • 09/2024 Joined Renmin University of China as an M.S. student in Computer Science.
Research
* denotes corresponding author
Recent works are highlighted.
Frame packing paper preview
Elephant in the Fridge: Constant-Memory Frame Packing for Long Video Understanding
Shuo Gao, Chenhao Zheng, Zhongzheng Ren*
In Neural Information Processing Systems (NeurIPS), 2026
Compositional zero-shot recognition paper preview
Do Multi-modal LLMs possess Compositional Zero-Shot Recognition Capabilities?
Yipeng Xu, Shuo Gao, Gang Yang*, Xinyue Ju, Yici Zhang, Ruixiang Zhao, Xirong Li
In International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2026
Multi-modal prototype optimization paper preview
Text-Driven Multi-Modal Prototype Optimization for Multi-Disease Recognition on Retinal Fundus Images
Shuo Gao, Gang Yang*, Yici Zhang, Jingjing Xu
In International Conference on Bioinformatics and Biomedicine (BIBM), 2025
Experience
I am deeply grateful to everyone who has hosted and mentored me for their generous support, guidance, and opportunities.
LinkerBot
July 2026 - Present
Research Intern
Hosts: Langyong Shi, Xiaoxiao Hu
Working on Embodied Foundation Models and World Action Models for Dexterous Manipulations.
RAIR Lab, UNC-Chapel Hill
February 2026 - Present
Research Intern (remote)
Research Advisor: Zhongzheng (Jason) Ren
Working on efficient VLMs for long video understanding.
360 AI Research
October 2025 - June 2026
Research Intern
Hosts: Jincheng Li, Chunyu Xie
Working on improving the localization capability of VLMs.
AIM Lab, University of Amsterdam
January 2025 - June 2025
Research Intern (remote)
Research Advisor: Yingjun Du
Working on Test-time Prompt Adaptation for VLMs.