S-Space: Exploring Spatial Workspace in Multimodal Models
Like MirroS-Lab/S-Space, 0 likes 0 Bookmark MirroS-Lab/S-Space, 0 bookmarks 0 TempoVista is the official implementation of Dyn-3D, a benchmark and training pipeline for 3D spatial reasoning in vision-language models. The release contains frame-selection tools, SFT and Kinematic-GSPO training code, benchmark evaluation scripts, Dyn-3D annotations, and checkpoints for Qwen3-VL and InternVL3.5.
Like InkMind-AI/Dyn-3D, 0 likes 0 Bookmark InkMind-AI/Dyn-3D, 0 bookmarks 0 Code repository for Qwen-3D: A Generalist 3D Vision-Language Model for Spatial Understanding
Like ll220/qwen3d, 0 likes 0 Bookmark ll220/qwen3d, 0 bookmarks 0 Clone of DeepSeek Thinking-with-Visual-Primitives
Like mitkox/Thinking-with-Visual-Primitives, 0 likes 0 Bookmark mitkox/Thinking-with-Visual-Primitives, 0 bookmarks 0 Learning Spatial Reasoning through Interaction with the Observable Physical World
Like ZJU-OmniAI/Spatial-Interactor, 0 likes 0 Bookmark ZJU-OmniAI/Spatial-Interactor, 0 bookmarks 0 [ICML 2026] ReVSI: Rebuilding Visual Spatial Intelligence Evaluation for Accurate Assessment of VLM 3D Reasoning
Like 3dlg-hcvc/revsi, 0 likes 0 Bookmark 3dlg-hcvc/revsi, 0 bookmarks 0 Puffin Series: Towards Unified Multimodal 3D World Models
Like KangLiao929/Puffin, 0 likes 0 Bookmark KangLiao929/Puffin, 0 bookmarks 0 Archived snapshot of Thinking-with-Visual-Primitives
Like ailuntx/Thinking-with-Visual-Primitives, 0 likes 0 Bookmark ailuntx/Thinking-with-Visual-Primitives, 0 bookmarks 0 Benchmark and training code for MindCube: spatial mental modeling in vision-language models from limited views.
Like mll-lab-nu/MindCube, 0 likes 0 Bookmark mll-lab-nu/MindCube, 0 bookmarks 0 
