Reconstruction as Spatial Evaluation: Diagnosing Cross-View Spatial Intelligence in Large Vision-Language Models
cs.CV, cs.AI
Submitted: 2026-11-01
Updated: 2026-11-01
Comments: 36 pages, 12 figures, including appendices
Code: https://github.com/Hyperwjf/ReconSpatial
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- GPT-4 Technical Report
- AgiBot World Colosseo: A Large-scale Manipulation Platform for Scalable and Intelligent Embodied Systems
- Qwen3-VL Technical Report
- nuScenes: A multimodal dataset for autonomous driving
- SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities
- BabyVision: Visual Reasoning Beyond Language
- Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
- ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes
- S-Agent: Spatial Tool-Use Elicits Reasoning for Spatial Intelligence
- EmbSpatial-Bench: Benchmarking Spatial Understanding for Embodied Tasks with Large Vision-Language Models
- Video-MME-v2: Towards the Next Stage in Benchmarks for Comprehensive Video Understanding
- BLINK: Multimodal Large Language Models Can See but Not Perceive
- Gemini Robotics: Bringing AI into the Physical World
- HallusionBench: An Advanced Diagnostic Suite for Entangled Language Hallucination and Visual Illusion in Large Vision-Language Models
- ESI-Bench: Towards Embodied Spatial Intelligence that Closes the Perception-Action Loop
- Video2Layout: Recall and Reconstruct Metric-Grounded Cognitive Map for Spatial Reasoning
- Galaxea Open-World Dataset and G0 Dual-System VLA Model
- A Diagram Is Worth A Dozen Images
- SpatialEvo: Self-Evolving Spatial Intelligence via Deterministic Geometric Environments
- ViewSpatial-Bench: Evaluating Multi-perspective Spatial Localization in Vision-Language Models
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs