Research map / Probabilistic and scientific modeling / Molecules and proteins
GeoWorld: Geometric World Models
Zeyu Zhang, Danning Li, Ian Reid, Richard Hartley
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| apply hyperbolic latent space (H-JEPA) plus geometric RL instead of Euclidean V-JEPA 2 | V-JEPA 2 ViT-g384 · CrossTask T=3 video goals | success rate (%): 50.16 -> 51.71 | Improved |
| apply GeoWorld instead of V-JEPA 2 | ViT-g384 · COIN T=4 image goals | success rate (%): 23.43 -> 27.79 | Improved |
| apply GeoWorld instead of V-JEPA 2 | ViT-L · CrossTask T=3 image goals | success rate (%): 43.33 -> 43.89 | No clear effect |
| apply GeoWorld instead of V-JEPA 2 | ViT-g384 · CrossTask T=4 image goals | success rate (%): 31.36 -> 31.48 | No clear effect |
| apply GeoWorld instead of V-JEPA 2 | ViT-g384 · CrossTask T=6 video goals | success rate (%): 16.88 -> 18.26 | Improved |
| apply GeoWorld instead of V-JEPA 2 | ViT-g384 · CrossTask video goals mIoU/mAcc | mIoU (%): COIN T=4 mIoU 94.51 -> 95.84; CrossTask T=3 mIoU 91.73 -> 92.95 | Mixed |
| apply GeoWorld instead of V-JEPA 2 on image-goal procedural planning | ViT-g384 · CrossTask T=3 | mIoU (%): 69.42 -> 86.55 | Improved |
Closest papers by meaning
- Predictive but Not Plannable: RC-aux for Latent World Models (NeurIPS 2026 · similarity 0.87)
- Sparse Planning in Visual World Models via Cost Gradients (NeurIPS 2026 · similarity 0.87)
- MindJourney: Test-Time Scaling with World Models for Spatial Reasoning (NeurIPS 2025 · similarity 0.87)
- Enhancing Geometric Perception in VLMs via Translator-Guided Reinforcement Learning (ICLR 2026 · similarity 0.86)
- MetaSpatial: Reinforcing 3D Spatial Reasoning in VLMs for the Metaverse (ICLR 2026 · similarity 0.86)
- Open-World Reinforcement Learning over Long Short-Term Imagination (ICLR 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
