AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

SpatialLogic-Bench: A Diagnostic Benchmark for Task-Oriented Spatiotemporal Reasoning

Xiaoda Yang, Shenzhou Gao, Can Wang, Jiahe Zhang, Menglan Tang, Jingyang Xue, Sheng Liu, Peijian Zhang, Yao Mu, Xiangyu Yue

AAAI 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Spatial reasoning in VLMs · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Introduces SpatialLogic-Bench, a benchmark for evaluating vision-language models on task-oriented spatiotemporal reasoning in dynamic scenes.

Problem

Existing benchmarks for vision-language models focus on static spatial relationships and do not systematically assess dynamic, task-oriented spatiotemporal reasoning.

Approach

A diagnostic benchmark targeting spatiotemporal logic in VLMs, aimed at dynamic task-oriented reasoning rather than static perception.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add ground-truth state-change analysis to prompt (ablation)GPT-4o · inverse task (window >=12)accuracy (%): 58.3 -> 95.2Improved
add ground-truth state-change analysis to prompt (ablation)GPT-o4mini · inverse task (window >=12)accuracy (%): 40.5 -> 98.6Improved
add ground-truth state-change analysis to prompt (ablation)Paligemma · inverse task (window >=12)accuracy (%): 8.3 -> 12.2No clear effect
add ground-truth state-change analysis to prompt (ablation)InternVL-8B · inverse task (window >=12)accuracy (%): 15.8 -> 21.5No clear effect
add ground-truth state-change analysis to prompt (ablation)DeepSeek-VL2 · inverse task (window >=12)accuracy (%): 69.0 -> 20.6Hurt
add ground-truth state-change analysis to prompt (ablation)Video-3D LLM · inverse task (window >=12)accuracy (%): 67.4 -> 78.5Improved
reverse frame order (inverse vs forward task)GPT-4o · SpatialLogic-Benchaccuracy (%): 99.8 -> 53.7Hurt
reverse frame order (inverse vs forward task)Paligemma · SpatialLogic-Benchaccuracy (%)Hurt
reverse frame order (inverse vs forward task)Gemini-2.5-Pro · SpatialLogic-Benchaccuracy (%): 82.7 -> 77.7No clear effect
increase window size between framesmost VLMs · SpatialLogic-Benchaccuracy (%): Gemini-2.5-Pro 71.3 -> 100.0Improved
increase window size between framesQwen2.5-VL-7B · SpatialLogic-Benchaccuracy (%): 53.1 -> 56.5No clear effect
use 2D open-source model instead of 3D-awareopen-source VLMs · SpatialLogic-Benchaccuracy (%): 3D-LLM 27.0 vs DeepSeek-VL2 64.4 (window 5)Improved

Experimental setup

Benchmarks: SpatialLogic-Bench

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