I’m Guanyu Zhou, a third-year undergraduate student interested in artificial intelligence. My current research interest is in multimodal large language models (MLLMs). Prior to this, I had an unforgettable research time at AIMlab, where I worked on problems related to action recognition and causal inference. I hope to learn from MLLMs, explore its bottlenecks, and explore the prototype of the next generation of intelligent systems. At this stage, I am committed to finding a more effective multimodal alignment solution. If you’re interested in collaborating with me on this topic, feel free to email me!😄

📅 Update

• [2024.11] The preprint of OccludeNet is now available, welcome to check it out!
• [2024.10] The preprint of CausalMM is now available, welcome to check it out!
• [2024.7] I am visiting HKUST(GZ) under the supervision of Professor Xuming Hu!

📝 Publications

(Preprint) Mitigating Modality Prior-Induced Hallucinations in Multimodal Large Language Models via Deciphering Attention Causality

Guanyu Zhou, Yibo Yan, Xin Zou, Kun Wang, Aiwei Liu, Xuming Hu

PDF Code

(Preprint) OccludeNet: A Causal Journey into Mixed-View Actor-Centric Video Action Recognition under Occlusions

Guanyu Zhou, Wenxuan Liu, Wenxin Huang, Xuemei Jia, Xian Zhong, Chia-Wen Lin

PDF Code

📇 Experience

[2024.7] HKUST(GZ) / AI Thrust

Bias in Model Prior Knowledge

[2023.7] AIMLab / Action Recogniton Group

Bias in the Data (Learning Process)

😘 Friends Links

• hanghang214 Outstanding infosec undergrad, admitted to master’s at PKU.