Yangguang Liu — accepted AI-conference papers

2026

BiSTAC: A Class of Bipedal Soft Terrestrial-Aerial Bicopters

Yangguang Liu, Yimin Zhu, Yue Wang, Ruoyu Ren, Lixian Zhang

RA-L 2026

This paper presents the design, modeling, and motion control of BiSTAC, a novel hybrid terrestrial–aerial vehicle. BiSTAC features an innovative configuration that integrates a bicopter propeller system with a flexible bipedal locomotion mechanism, thereby enabling both bicopter flight and bipedal w

Cited by 0 Source ↗Scholar ↗

2026

Parameter-Masked Decoupled Optimization for Cross-Domain Class-Incremental Learning

Ziqi Gu, Yangguang Liu, Wenxuan Fang, Baotong Su, Dan Wang, Tong Zhang +2

ICML 2026poster

Cross-domain class-incremental learning (CD-CIL) requires models to continuously acquire new classes across shifting domains while retaining previously learned knowledge. Existing approaches often entangle what to update with how to update, resulting in unstable adaptation and severe forgetting unde…

Cited by 0 Source ↗Scholar ↗