Egocentric Instruction-oriented Affordance Prediction via Large Multimodal Model
Bokai Ji, Jie Gu, Xiaokang Ma, Chu Tang, Jingmin Chen, Guangxia Li
Under Review
Research
Our research focuses on embodied AI, multimodal learning, and robotics. Selected publications and preprints are listed below.
Generic Token Compression in Multimodal Large Language Models from an Explainability Perspective
Lei Lei, Jie Gu, Xiaokang Ma, Chu Tang, Jingmin Chen, Tong Xu
Under Review
Stimulating Imagination: Towards General-purpose "Something Something Placement"
Jianyang Wu, Jie Gu, Xiaokang Ma, Fangzhou Qiu, Chu Tang, Jingmin Chen
IROS 2025Accepted
Gaussian Sequences: Self-bootstrapping Dynamic Gaussian Reconstruction from Monocular Casual Videos
Under Review
QCDT: Efficiently Optimize and Deploy Detection Transformers via Query Compression
Under Review