AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

pySpatial: Generating 3D Visual Programs for Zero-Shot Spatial Reasoning

Zhanpeng Luo, Ce Zhang, Silong Yong, Cunxi Dai, Qianwei Wang, Haoxi Ran, Guanya Shi, Katia P. Sycara, Yaqi Xie

ICLR 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Open-vocabulary alignment from VLMs · 10 citations (OpenAlex)

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 changedWhereMeasuredEffect
add pySpatial 3D visual programming to code agentGPT-4.1-mini · MindCube (full)overall accuracy: 45.62 -> 58.56Improved
add pySpatial (ablation)GPT-4o · MindCube-1koverall accuracy: 42.29 -> 62.67Improved
add pySpatial (ablation)GPT-4.1 · MindCube-1koverall accuracy: 44.67 -> 62.35Improved
add pySpatial (single-view)GPT-4o · Omni3D-Benchtotal score: 42.9 -> 44.2Improved
add pySpatialGPT-4o · MMSI-Benchaverage accuracy: 30.3 -> 37.3Improved
use explicit 3D reconstruction tools instead of implicit mental models (CoT, cognitive map)Qwen2.5-VL-3B mental models vs pySpatial · MindCube-1koverall accuracy: 41.43 -> 62.35Improved
give VADAR the 3D reconstruction module (ablation)VADAR · MindCube-1koverall accuracy: 40.76 -> 35.62Hurt
add in-context examples to code agent (0 -> 2) (ablation)pySpatial GPT-4o · MindCube-1koverall accuracy: 53.62 -> 62.67Improved
increase in-context examples from 2 to 4 (ablation)pySpatial GPT-4o · MindCube-1koverall accuracy: 62.67 -> 63.14No clear effect
swap 3D reconstruction backbone (VGGT, Pi3, CUT3R) (ablation)pySpatial GPT-4o · MindCube-1koverall accuracy: 62.67 / 63.33 / 61.05No clear effect
swap code agent (GPT-4o, Qwen3-Coder, DeepSeek-v3) (ablation)pySpatial · MindCube-1koverall accuracy: 62.67 / 62.10 / 61.05No clear effect
use pySpatial on Rotation subset with GPT-4.1-mini agent vs Claude-4-SonnetpySpatial · MindCube (full) Rotationaccuracy: 48.42 (Claude-4-Sonnet) vs 43.20Hurt

Later papers that compared with this method

2 later accepted papers report pySpatial as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy