🚀 Introducing World-R1: Video models already know 3D — they just need RL to wake it up! No arch changes. No video training data. No extra inference cost.⬇️ 🌐Website: aka.ms/world-r1

📢 We release "Feed-Forward 3D Scene Modeling: A Problem-Driven Perspective" — a comprehensive survey covering 200+ papers on feed-forward 3D reconstruction! Instead of categorizing by 3D representations, we propose a problem-driven taxonomy. 🌐 ff3d-survey.github.io

We are excited to share our latest work: CoV: Chain-of-View Prompting for Spatial Reasoning. CoV is a multi-step reasoning Embodied QA agent framework that enables VLMs to actively acquire more question-relevant viewpoints. Project on ziplab.co/CoV/

VolSplat: Rethinking Feed-Forward 3D Gaussian Splatting with Voxel-Aligned Prediction @weijiewang.bsky.social, Yeqing Chen, Zeyu Zhang, Hengyu Liu, Haoxiao Wang, Zhiyuan Feng, Wenkang Qin, Zheng Zhu, @donydchen.bsky.social, Bohan Zhuang arxiv.org/abs/2509.19297

𝗩𝗼𝗹𝗦𝗽𝗹𝗮𝘁: 𝗥𝗲𝘁𝗵𝗶𝗻𝗸𝗶𝗻𝗴 𝗙𝗲𝗲𝗱-𝗙𝗼𝗿𝘄𝗮𝗿𝗱 𝟯𝗗 𝗚𝗮𝘂𝘀𝘀𝗶𝗮𝗻 𝗦𝗽𝗹𝗮𝘁𝘁𝗶𝗻𝗴 𝘄𝗶𝘁𝗵 𝗩𝗼𝘅𝗲𝗹-𝗔𝗹𝗶𝗴𝗻𝗲𝗱 𝗣𝗿𝗲𝗱𝗶𝗰𝘁𝗶𝗼𝗻 Weijie Wang, Yeqing Chen, Zeyu Zhang ... Bohan Zhuang arxiv.org/abs/2509.19297 Trending on www.scholar-inbox.com

ZPressor: Bottleneck-Aware Compression for Scalable Feed-Forward 3DGS @weijiewang.bsky.social, @donydchen.bsky.social, Zeyu Zhang, Duochao Shi, Akide Liu, Bohan Zhuang tl;dr: views->anchor & support sets; support view info->anchor views->compressed latent state Z arxiv.org/abs/2505.23734

🚀 We're excited to introduce ZPressor, a bottleneck-aware compression module for scalable feed-forward 3DGS. Plug & play, lightweight, and powerful. Read more: lhmd.top/zpressor/ Paper: www.arxiv.org/abs/2505.23734 Code: github.com/ziplab/ZPres...