RoboSPA: Can VLA Models Go Beyond Simple Scenes and Short-Horizon Tasks?
RoboSPA introduces a 527K-trajectory robotic benchmark with 280 task variants exposing VLA model weaknesses in spatial reasoning and long-horizon planning.
RoboSPA is a large-scale robotic manipulation dataset and benchmark for diagnosing embodied reasoning in vision-language-action models, spanning fine-grained spatial reasoning and long-horizon procedural planning. It covers 10 task categories and 56 base tasks instantiated across five difficulty levels for 280 variants, with 527K trajectories collected across multiple embodiments. Experiments on representative VLA models show current systems struggle with complex spatial relations, precise low-level execution, and memory-intensive planning.
- 280 task variants across five difficulty levels
- 527K trajectories across multiple embodiments and scenes
- Introduces diagnostic metrics beyond binary success rate
- Current VLA models struggle with spatial relations and long-horizon planning
From the original article
1 min read at arxiv.orgVision-Language-Action (VLA) models have shown promising progress in language-conditioned robotic manipulation. However, existing datasets and benchmarks mainly evaluate task completion under predefined settings, offering limited insight into model reasoning under increasing spatial and procedural complexity.
By Zhenxuan Fan · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.