# GST-Bench: Can VLMs Develop Global Spatial Awareness from Video?

arXiv: https://arxiv.org/abs/2608.05747 · HTML: https://mlsanity.com/paper/2608.05747/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 30 references that are in the map.
- First posted on arXiv 6 Aug 2026.

## Builds on (most cited first, 27 of 27)

- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency](https://mlsanity.com/paper/2508.18265/) (2025)
- [OpenAI GPT-5 System Card](https://mlsanity.com/paper/2601.03267/) (2025)
- [Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs](https://mlsanity.com/paper/2406.16860/) (2024)
- [Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces](https://mlsanity.com/paper/2412.14171/) (2024)
- [Hypersim: A Photorealistic Synthetic Dataset for Holistic Indoor Scene Understanding](https://mlsanity.com/paper/2011.02523/) (2020)
- [BLINK: Multimodal Large Language Models Can See but Not Perceive](https://mlsanity.com/paper/2404.12390/) (2024)
- [Visual Spatial Reasoning](https://mlsanity.com/paper/2205.00363/) (2022)
- [What's "up" with vision-language models? Investigating their struggle with spatial reasoning](https://mlsanity.com/paper/2310.19785/) (2023)
- [LLaVA-OneVision-1.5: Fully Open Framework for Democratized Multimodal Training](https://mlsanity.com/paper/2509.23661/) (2025)
- [NVILA: Efficient Frontier Visual Language Models](https://mlsanity.com/paper/2412.04468/) (2024)
- [MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence](https://mlsanity.com/paper/2505.23764/) (2025)
- [3DSRBench: A Comprehensive 3D Spatial Reasoning Benchmark](https://mlsanity.com/paper/2412.07825/) (2024)
- [From Flatland to Space: Teaching Vision-Language Models to Perceive and Reason in 3D](https://mlsanity.com/paper/2503.22976/) (2025)
- [Cambrian-S: Towards Spatial Supersensing in Video](https://mlsanity.com/paper/2511.04670/) (2025)
- [ViewSpatial-Bench: Evaluating Multi-perspective Spatial Localization in Vision-Language Models](https://mlsanity.com/paper/2505.21500/) (2025)
- [STI-Bench: Are MLLMs Ready for Precise Spatial-Temporal World Understanding?](https://mlsanity.com/paper/2503.23765/) (2025)
- [MM-Spatial: Exploring 3D Spatial Understanding in Multimodal LLMs](https://mlsanity.com/paper/2503.13111/) (2025)
- [Seed1.8 Model Card: Towards Generalized Real-World Agency](https://mlsanity.com/paper/2603.20633/) (2026)
- [RoboBrain 2.5: Depth in Sight, Time in Mind](https://mlsanity.com/paper/2601.14352/) (2026)
- [Spatial457: A Diagnostic Benchmark for 6D Spatial Reasoning of Large Multimodal Models](https://mlsanity.com/paper/2502.08636/) (2025)
- [MMSI-Video-Bench: A Holistic Benchmark for Video-Based Spatial Intelligence](https://mlsanity.com/paper/2512.10863/) (2025)
- [OST-Bench: Evaluating the Capabilities of MLLMs in Online Spatio-temporal Scene Understanding](https://mlsanity.com/paper/2507.07984/) (2025)
- [Robix: A Unified Model for Robot Interaction, Reasoning and Planning](https://mlsanity.com/paper/2509.01106/) (2025)
- [ArtVIP: Articulated Digital Assets of Visual Realism, Modular Interaction, and Physical Fidelity for Robot Learning](https://mlsanity.com/paper/2506.04941/) (2025)

## Abstract (arXiv)

Spatial intelligence is fundamental to embodied agents, yet existing benchmarks focus on local spatial perception from single or few viewpoints, overlooking global spatial awareness over continuous, long-horizon visual streams. To address this limitation, we introduce the Global-Spatial-Temporal Benchmark (GST-Bench), a VQA benchmark for global spatial intelligence in video understanding, comprising human-verified questions derived from 6,790 minutes of synthetically generated video. It requires models to perform accurate spatial inference from novel viewpoints unseen in the input video and to map egocentric observations onto global top-down images. A comprehensive evaluation of 22 state-of-the-art VLMs exposes a striking gap between models and humans: the strongest zero-shot model attains only 42.68, far below the human score of 79.08. To probe the cause of this gap, we construct GST-Bench-Local and find that models, despite strong local spatial understanding under the same task formulation, still fail to consolidate long-horizon observations into a globally consistent scene representation. We further provide GST-Train, a dataset for global spatial reasoning, as a complementary resource to facilitate future research on this challenge.