Research map / 3D vision / 3D generation
Learning Interpretable Spatial Operations in a Rich 3D Blocks World
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| use the proposed interpretable model instead of Bisk 16 | Blocks v1 · Blocks v1 | end-to-end target error (mean, block-lengths): 0.98 -> 0.80 | Helped |
| use the proposed interpretable model instead of Pišl 17 | Blocks v1 · Blocks v1 | end-to-end target error (mean, block-lengths): 0.72 -> 0.80 | Hurt |
| evaluate on the richer 3D Blocks v2 corpus rather than Blocks v1 | Ours · Blocks v1 versus Blocks v2 | end-to-end target error (mean, block-lengths): 0.80 -> 1.15 | Hurt |
| train jointly on v1 and v2 | Ours, evaluated on Blocks v1 · Blocks v1 | end-to-end target error (mean, block-lengths): 0.80 -> 0.84 | No clear effect |
| train jointly on v1 and v2 | Ours, evaluated on Blocks v2 · Blocks v2 | end-to-end target error (mean, block-lengths): 1.15 -> 1.35 | Hurt |
| train jointly on v1 and v2 | Ours, evaluated on Blocks v1 · Blocks v1 | source accuracy: 97.5 -> 98.1 | Helped |
| train jointly on v1 and v2 | Ours, evaluated on Blocks v2 · Blocks v2 | source accuracy: 91.3 -> 93.1 | Helped |
| predict target location with the correct source block provided | Ours on Blocks v1 · Blocks v1 | target error (mean, block-lengths): 0.80 -> 0.7 | Helped |
Closest papers by meaning
- Surprise3D: A Dataset for Spatial Understanding and Reasoning in Complex 3D Scenes (NeurIPS 2025 · similarity 0.86)
- From Flatland to Space: Teaching Vision-Language Models to Perceive and Reason in 3D (NeurIPS 2025 · similarity 0.86)
- Spatial Understanding from Videos: Structured Prompts Meet Simulation Data (NeurIPS 2025 · similarity 0.86)
- SpatialStack: Layered Geometry-Language Fusion for 3D VLM Spatial Reasoning (CVPR 2026 · similarity 0.85)
- SPARTUN3D: Situated Spatial Understanding of 3D World in Large Language Model (ICLR 2025 · similarity 0.85)
- CitySTAR: Agent-Driven Structured and Topology-Aware Reasoning for Open-Vocabulary Urban 3D Grounding (NeurIPS 2026 · similarity 0.85)
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.