10 papers
Decoupled Drift-Diffusion World Models for Consistent Future Frame Prediction
Anton Arapin, Aditya Singh, Yixin Wang +2
World models of visual information predict future image frames conditioned on given action sequences. This predicted scene evolution is driven by deterministic changes such as cont…
MeshPriorDiT: Hierarchical Modeling for Action-Conditioned Cloth Dynamics
Zihang Wang, Jianming Hu, Shang Su +4
Action-conditioned cloth dynamics prediction requires both locally plausible deformation and long-range coordination. Existing approaches largely follow two paradigms. Mesh-based G…
ReOrder-OPD:Reliability-Aware Prompt Ordering for On-Policy Distillation
Ximo Zhu, Ruiqi Liu, Rong Wang +8
On-policy distillation (OPD) applies token-level teacher supervision to student-generated trajectories, but this supervision is not always reliable. Existing methods use local conf…
DC-WAM: Dynamic-Centric Visual Supervision and Reasoning for World-Action Models
Haoyuan Ji, Lingxiang Fan, Shang Su +4
World-Action Models (WAMs) augment robot policies with future visual prediction, but it remains unclear what the visual modality should learn for control. While photorealistic futu…
: 3D Reconstruction via Relative Regression
Congrong Xu, Huachen Gao, Xingyu Chen +3
Recent feed-forward geometry foundation models have demonstrated impressive generalization by recovering depth and poses in a single forward pass. However, these models are typical…
SGNN: Efficient Global Mixing and Local Message Passing for Long-Range Graph Learning
Dai Shi, Luke Thompson, Linhan Luo +4
Message-passing neural networks (MPNNs) often suffer from an information bottleneck when capturing long-range dependencies, leading to the oversquashing (OSQ) phenomenon. Alongside…
Visual-Advantage On-Policy Distillation for Vision-Language Models
Ruiqi Liu, Xiaolei Lv, Gengsheng Li +8
On-policy knowledge distillation has proven effective for language models, yet its application to vision-language models (VLMs) remains underexplored. We observe that standard on-p…
Lyra 2.0: Explorable Generative 3D Worlds
Tianchang Shen, Sherwin Bahmani, Kai He +12
Recent advances in video generation enable a new paradigm for 3D scene creation: generating camera-controlled videos that simulate scene walkthroughs, then lifting them to 3D via f…
The Dynamic Prior: Understanding 3D Structures for Casual Dynamic Videos
Zhuoyuan Wu, Xurui Yang, Jiahui Huang +2
Estimating accurate camera poses, 3D scene geometry, and object motion from in-the-wild videos is a long-standing challenge for classical structure from motion pipelines due to the…
Lyra: Generative 3D Scene Reconstruction via Video Diffusion Model Self-Distillation
Sherwin Bahmani, Tianchang Shen, Jiawei Ren +10
The ability to generate virtual environments is crucial for applications ranging from gaming to physical AI domains such as robotics, autonomous driving, and industrial AI. Current…