6 papers
PhysDrift: Bridging the Embodiment Gap in Humanoid Co-Speech Motion Generation
Zhangzhao Liang, Xiaofen Xing, Mingyue Yang +2
Humanoid robots require co-speech motions that are not only expressive and speech-aligned, but also physically executable under embodiment constraints. Existing co-speech generatio…
Motion-Adapter: A Diffusion Model Adapter for Text-to-Motion Generation of Compound Actions
Yue Jiang, Mingyu Yang, Liuyuxin Yang +3
Recent advances in generative motion synthesis have enabled the production of realistic human motions from diverse input modalities. However, synthesizing compound actions from tex…
Retrieval Feedback Memory Enhancement Large Model Retrieval Generation Method
Leqian Li, Dianxi Shi, Jialu Zhou +4
Large Language Models (LLMs) have shown remarkable capabilities across diverse tasks, yet they face inherent limitations such as constrained parametric knowledge and high retrainin…
CEIDM: A Controlled Entity and Interaction Diffusion Model for Enhanced Text-to-Image Generation
Mingyue Yang, Dianxi Shi, Jialu Zhou +4
In Text-to-Image (T2I) generation, the complexity of entities and their intricate interactions pose a significant challenge for T2I method based on diffusion model: how to effectiv…
Dynamic Embedding of Hierarchical Visual Features for Efficient Vision-Language Fine-Tuning
Xinyu Wei, Guoli Yang, Jialu Zhou +4
Large Vision-Language Models (LVLMs) commonly follow a paradigm that projects visual features and then concatenates them with text tokens to form a unified sequence input for Large…
Separation and Collaboration: Two-Level Routing Grouped Mixture-of-Experts for Multi-Domain Continual Learning
Jialu Zhou, Dianxi Shi, Shaowu Yang +5
Multi-Domain Continual Learning (MDCL) acquires knowledge from sequential tasks with shifting class sets and distribution. Despite the Parameter-Efficient Fine-Tuning (PEFT) method…