Yan: Foundational Interactive Video Generation
Deheng Ye, Fangyun Zhou, Jiacheng Lv, Jianqi Ma, Jun Zhang, Junyan Lv, Junyou Li, Minwen Deng, Mingyu Yang, Qiang Fu, Wei Yang, Wenkai Lv, Yangbin Yu, Yewen Wang, Yonghang Guan, Zhihao Hu, Zhongbin Fang, Zhongqian Sun. (listed in alphabetical order)

• Yan is a foundational framework for interactive video generation, featuring Yan-Sim for high-quality simulation, Yan-Gen for generating videos from text and images, and Yan-Edit for real-time editing, tested using games to ensure high performance and visual quality.

Human-in-Context: Unified Cross-Domain 3D Human Motion Modeling via In-Context Learning
Mengyuan Liu, Xinshun Wang📧, Zhongbin Fang📧, Deheng Ye, Xia Li, Tao Tang, Songtao Wu, Xiangtai Li, Ming-Hsuan Yang

• An in-context cross-domain model in 3D human motion modeling that simultaneously possesses cross-modality,cross task, and cross-dataset generalization capability.

Playable Game Generation
Mingyu Yang, Junyou Li, Zhongbin Fang, Sheng Chen, Yangbin Yu, Qiang Fu, Wei Yang, Deheng Ye.

• We develop PlayGen to enable playable game generation with real-time interaction, sufficient visual quality, and accurate simulation of interactive mechanics.

Point-In-Context: Understanding Point Cloud via In-Context Learning
Mengyuan Liu, Zhongbin Fang📧, Xia Li📧, Joachim M Buhmann, Deheng Ye, Xiangtai Li, Chen Change Loy

• Possess stronger performance and generalization ability and achieve SOTA in multi-dataset segmentation tasks.
• Can perform unique part segmentation tasks via customized prompts.

Skeleton-in-Context: Unified Skeleton Sequence Modeling with In-Context Learning
Xinshun Wang*, Zhongbin Fang*, Xia Li, Xiangtai Li, Chen Chen, Mengyuan Liu📧

• The first to explore in-context learning for skeleton sequence modeling.
• Solve the over-fitting problem encountered when using the masked-modeling-style training framework of existing methods.

Explore In-Context Learning for 3D Point Cloud Understanding (Spotlight)
Zhongbin Fang, Xiangtai Li, Xia Li, Joachim M Buhmann, Chen Change Loy, Mengyuan Liu📧

• The first to explore in-context learning for 3D point cloud.
• Solve the information leakage problem caused by previous MPM methods.
• Can improve the performance by selecting higher-quality prompts

ModelNet-O: A Large-Scale Synthetic Dataset for Occlusion-Aware Point Cloud Classification
Zhongbin Fang, Xia Li, Xiangtai Li, Shen Zhao📧, Mengyuan Liu📧

• Propose a challenging occluded point cloud classification dataset ModelNet-O.
• Propose a robust occlusion-aware method, PointMLS, based on a multi-level sampling strategy.