

Research Interests
The Machine Perception and Learning (MAPLE) Lab has long been engaged in artificial intelligence, especially deep learning and its research in generative Artificial intelligence (AIGC), multimodal understanding and generative large models, 2D/3D intelligent creation, virtual reality, etc. 1.Intelligent creation based on generative artificial intelligence, such as GAN model, diffusion model, self-coding model and other deep learning theory and application;2.Multi-modal understanding and generation direction: video and image multi-modal generation and editing, 2D/3D scene and human body perception, reconstruction and interaction;3.Theory and application of artificial intelligence in 2D/3D virtual content creation, especially automatic 2D/3D virtual reality object and scene generation, non-rigid object generation and driving model with human main object;4.Embodied intelligence, including the design of general-purpose vision-language-action (VLA) models, robotic manipulation with robotic arms and dexterous hands, embodied agent reasoning, and world model–based VLA learning;
Group News
-
Sep 19, 2025 Accepted by NeurIPS 2025
Our Paper "Reinforcing the Diffusion Chain of Lateral Thought with Diffusion Language Models" has been accepted by NeurIPS 2025. Congratulations to MAPLE Group!
-
Feb 27, 2025 Accepted by CVPR 2025
Our Paper "Schedule On the Fly: Diffusion Time Prediction for Faster and Better Image Generation" has been accepted by CVPR 2025. Congratulations to MAPLE Group!
-
Dec 16, 2024 Accepted by ACM Multimedia 2024
Our Paper on “Equilibrated Diffusion: Frequency-aware Textual Embedding for Equilibrated Image Customization” was accepted by ACM Multimedia 2024 as oral project, where we demonstrate that Equilibrated Diffusion surpasses other competitors with better subject consistency while closely adhering to text descriptions, thus validating the superiority of our approach