Publications

* Equal contribution. ♡ Project lead. ✉ corresponding/co-corresponding author.
Also see my Google Scholar profile for the most up-to-date list.

2026

  • children_gait
    Decoding Children's Gait Behavior
    Yifan Shen*, Boyi Li*, Meihuan Huang*, Yuanzhe Liu*, Xu Cao*,♡, Jinyang Jin, Zhengyuan Li, Anglin Liu, Junho Kim, Jingyuan Zhu, Lan Fangzhou, Jianguo Cao, Jintai Chen, Ismini Lourentzou, James M. Rehg✉
    Paper Code Website
    The first and largest repository of multi-camera gait videos of children with developmental motor conditions, containing both clinically relevant assessments (EVGS and GVS) and rich frame-level annotations. The first comprehensive assessment of the ability of SoTA open and closed VLMs to decode the subtle signs of gait abnormalities from video, and find that out-of-the-box models are ineffective for this task. A novel video analysis method that is the current SoTA for the automated inference of EVGS and GVS scores from children's videos.
    ECCV 2026

  • Challenge
    The 1st AI Children Challenge
    Boyi Li, Yifan Shen, Houze Yang, Xu Cao, Guojun Yun, Li Gao, Turong Chen, Long Xu, Jianguo Cao, Meihuan Huang✉
    Paper
    The 1st AI Children Challenge aims to advance real-world applications of computer vision and AI in child healthcare, child education, and pediatrics. The 2026 CV4CHL edition featured the first track in this domain: Children Gait Visual Analysis. The main goal of Children Gait Visual Analysis is the fine-grained analysis of children's gait behaviors from keypoint sequences.
    CVPR 2026 Workshop on Computer Vision for Children

  • Cognitive
    Toward Cognitive Supersensing in Multimodal Large Language Model
    Boyi Li*, Yifan Shen*,♡, Yuanzhe Liu*, Yifan Xu, Jiateng Liu, Xinzhuo Li, Zhengyuan Li, Jingyuan Zhu, Yunhan Zhong, Fangzhou Lan, Jianguo Cao, James M. Rehg, Heng Ji, Ismini Lourentzou✉, Xu Cao✉
    Paper Code Website
    A novel training paradigm that endows MLLMs with human-like visual imagery capabilities by integrating a Latent Visual Imagery Prediction (LVIP) head that jointly learns sequences of visual cognitive latent embeddings and aligns them with the answer, thereby forming vision-based internal reasoning chains. To evaluate the cognitive capabilities of MLLMs, we present CogSense-Bench, a comprehensive visual question answering (VQA) benchmark assessing five cognitive dimensions.
    arXiv Preprint

2025

  • AGP
    Adaptive Graph Pruning for Multi-Agent Communication
    Boyi Li*, Zhonghan Zhao*,♡, Der-Horng Lee✉, Gaoang Wang✉
    Paper Code Website
    A novel task-adaptive multi-agent collaboration framework that jointly optimizes agent quantity (hard-pruning) and communication topology (soft-pruning), dynamically constructing optimized communication topologies tailored specifically to individual tasks.
    ECAI 2025

2024

  • STEVE
    See and think: Embodied agent in virtual environment
    Zhonghan Zhao*, Wenhao Chai*,♡, Xuan Wang*, Boyi Li, Shengyu Hao, Shidong Cao, Tian Ye, Jenq-Neng Hwang, Gaoang Wang✉
    Paper Code Website
    A comprehensive and visionary embodied agent in the Minecraft virtual environment comprises three key components: vision perception, language instruction, and code action.
    ECCV 2024