Note: The program is subject to change. Final details will be updated closer to the workshop date.
The program schedule for the workshop is tentatively set as follows:
| Time | Event | Speaker/Presenter |
|---|---|---|
| 09:00 - 09:15 | Opening Remarks | Workshop Chairs |
| 09:15 - 10:05 | Keynote | Liang Zheng (Australian National University) |
| 10:05 - 10:55 | Keynote | Liangliang Cao (Hong Kong Polytechnic University) |
| 10:55 - 11:10 | Coffee Break | - |
| 11:10 - 12:00 | Keynote | Mike Shou (National University of Singapore) |
| 12:00- 12:30 | Challenge Track: "M3ISR: A Multi-Modal Multi-View Benchmark for 3D/4D Gaussian Splatting and Feedforward Compression" | Xinhui Liu (The University of Hong Kong) |
| 12:30 - 13:30 | Lunch Break | - |
| 13:30 - 13:45 | Challenge Track: "Predict-and-Verify Rendering for Dynamic Scenes with Semantic 4D Gaussians" | Lebin Zhou (Santa Clara University) |
| 13:45 - 14:35 | Keynote | Jinwei Gu (Nvidia) |
| 14:35 - 15:25 | Keynote | Junsong Yuan (SUNY Buffalo) |
| 15:25 - 15:45 | Coffee Break | - |
| 15:45 - 16:00 | Paper: "Reliability-Aware Intervention Policies for Interactive Gaussian SLAM" | Ivan Malashin (Bauman Moscow State Technical University) |
| 16:00 - 16:15 | Paper: "The RenAIssance: Caption-Guided Diffusion Models for Historical-Art Visual Media" | Zhanyu Tuo (Sorbonne Université) |
| 16:15 - 16:30 | Paper: "Chinese Composite Character Generation via Recursive IDS-Tree Processing and Component-Level Style Transfer" | Lingqi Li (Beijing Electronic Science and Technology Institute) |
| 16:30 - 16:45 | Paper: "Semantic World Priors for Training-Free Cross-Clip Consistency in Black-Box Video Generation" | Fuyuan Li (Amazon) |
| 16:45 - 17:00 | Paper: "Mechanistic Interpretability of Structure-Aware Numerical Reasoning in LLaMA 3.1-8Br" | Rahul Chowdhury (Northeastern University) |
| 17:00 - 17:15 | Paper: "AesCAST-An Automated Framework for Aesthetically-Guided Style Transfer" | Jiangqing Ye (Beijing Electronic Science and Technology Institute) |
| 17:15 - 17:30 | Paper: "TBQ:Trajectory-Aligned Branch-Aware Quantization for Few-Step MoE-Like Video Diffusion Models" | Jinyang Du (BUAA) |
| 17:30 - 17:45 | Paper: "Direct, Parallel, or Sequential? A Comparative Study of Training-Free Multi-Subject Image-to-Video Generation" | Yanliang Qi (The University of Memphis) |
| 17:45 - 18:00 | Award Session | Award Winners |
| 18:00 - 18:15 | Closing Remarks | Workshop Chairs |