5 papers
Decoupling Intention from Trajectory: A Representational Deduction Framework for World Action Models
Xiangkai Ma, Yue Ma, Junjie Wang +6
World Action Models (WAMs) aim to construct a unified architecture capable of understanding world state evolution and guiding to generative motion planning. However, existing visua…
S2Accompanist: A Semantic-Aware and Structure-Guided Diffusion Model for Music Accompaniment Generation
Huakang Chen, Wenkai Cheng, Guobin Ma +7
High-fidelity text-to-music generation typically relies on massive proprietary datasets and immense computational resources. Existing models often struggle to generate coherent pur…
Unifying Perception and Action: A Hybrid-Modality Pipeline with Implicit Visual Chain-of-Thought for Robotic Action Generation
Xiangkai Ma, Lekai Xing, Han Zhang +2
Vision-Language-Action (VLA) models built upon Chain-of-Thought (CoT) have achieved remarkable success in advancing general-purpose robotic agents, owing to its significant percept…
Temporal-Visual Semantic Alignment: A Unified Architecture for Transferring Spatial Priors from Vision Models to Zero-Shot Temporal Tasks
Xiangkai Ma, Han Zhang, Wenzhong Li +1
Large Multimodal Models (LMMs) have achieved remarkable progress in aligning and generating content across text and image modalities. However, the potential of using non-visual, co…
Domain Fusion Controllable Generalization for Cross-Domain Time Series Forecasting from Multi-Domain Integrated Distribution
Xiangkai Ma, Xiaobin Hong, Mingkai Lin +3
Conventional deep models have achieved unprecedented success in time series forecasting. However, facing the challenge of cross-domain generalization, existing studies utilize stat…