研究库 论文知识库

内容库 / 论文

arXiv:2609.05324 · 多模态

RoboSPA: Can VLA Models Go Beyond Simple Scenes and Short-Horizon Tasks?

RoboSPA: Can VLA Models Go Beyond Simple Scenes and Short-Horizon Tasks?

类型：arxiv
标识：2609.05324
链接：https://arxiv.org/abs/2609.05324
主分类：multimodal
形态：benchmark
TLDR：Vision-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. We introduce RoboSPA (Robot Spatial-Procedural Assessment), a large-scale robotic manipulation dataset and benchmark for diagnosing embodied reasoning in VLA models. RoboSPA focuses on two core dimensions, Fine-Grained Spatial Reasoning and Long-Horizon Procedural Planning, covering 10 ta
副分类：evaluation
待LLM分类：否
来源文件：
/inbox/tom/_candidates/2026-09-09-agent-rag-longcontext-candidates.json

相关 · 按语义匹配

笔记 Kimi K2.5: Visual Agentic Intelligence · 精读与批判

笔记 Tom 文献雷达 · AI Agent × RAG × 长上下文 × 评测

笔记 M3-Agent + M3-Bench · 短审稿（2026-08-11 15:50 · flyP 精读棒）

笔记 Tom 文献雷达 · AI Agent / RAG / 长上下文 · 2026-08-20

笔记 Cameron R. Wolfe (Deep Learning Focus) · RSS 摘要

仓库 trololollo78/Prompt-Engineering

arXiv

2609.05324

TLDR

Vision-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. We introduce RoboSPA (Robot Spatial-Procedural Assessment), a large-scale robotic manipulation dataset and benchmark for diagnosing embodied reasoning in VLA models. RoboSPA focuses on two core dimensions, Fine-Grained Spatial Reasoning and Long-Horizon Procedural Planning, covering 10 ta

arXiv ↗ 中文全文