Lists (1)
Starred repositories
[ICLR 2025] SPA: 3D Spatial-Awareness Enables Effective Embodied Representation
About This repository is a curated collection of the most exciting and influential CVPR 2026 papers. 🔥 [Paper + Code + Demo]
World Model Self-Distillation project website
Official implementation of MaskWAM: Unifying Mask Prompting and Prediction for World-Action Models
Official implementation of Surflo: Consistent 3D Surface Flow Model with Global State.
Office inference code for World Tracing (object/scene/dynamic). Live demos: https://haoz19.github.io/world-tracing-page/
Official code release for Reroute, Don’t Remove: Recoverable Visual Token Routing for Vision-Language Models.
[ICML 2026] PyTorch implementation of BudCache
Flex4DHuman turns monocular or sparse multi-view videos of dynamic subjects into synchronized dense multi-view videos.
Code for RepWAM: World Action Modeling with Representation Visual-Action Tokenizers
Envision4D: Envisioning Visual Futures via Feed-forward 4D Gaussian Splatting for Autonomous Driving
WorldOlympiad: Can Your World Model Survive a Triathlon?
Benchmarking MLLMs for Parametric 3D Generation and Structural Reasoning (Text-to-3D, Image-to-3D, Assembly-3D)
[CVPR 2025🔥] Official codebase for "Global-Local Tree Search in VLMs for 3D Indoor Scene Generation" and our arxiv 2026 extension
Official Pytorch implementation of the paper: "SAM-Flow: Source Anchored Masked Flow for Training-Free Image Editing"
Official Implementations of "RhymeFlow: Training-Free Acceleration for Video Generation with Asynchronous Denoising Flow Scheduling"
Official implementation of Complexity-Balanced Diffusion Splitting
Official PyTorch implementation of the paper "GP-Adapter: Gaussian Process CLIP-Adapter for Few-Shot OOD Detection", IJCNN 2026
Prisma-World: Camera-Controllable Multi-Agent Video World Model
[ICML 2026] SSR-Merge: Subspace Signal Routing for Training-Free LoRA Merging in Diffusion Models
[ICML 2026] The official implementation of "Mean Flow Distillation: Robust and Stable Distillation for Flow Matching Models".
Segment Anything Model for Medical Image Segmentation: Open-Source Project Summary
The code for the paper "LCM: Locally Constrained Compact Point Cloud Model for Masked Point Modeling" (NeurIPS'24).
[NeurIPS 2025] Mamba Goes HoME: Hierarchical Mixture-of-Experts for 3D Medical Image Segmentation