SpatialBench: Is Your Spatial Foundation Model an All-Round Player?

Hugging Face Daily Papers Papers
spatial-foundation-models benchmark evaluation dataset computer-vision representation-learning 3d-vision

Summary

SpatialBench is a comprehensive benchmark for evaluating spatial foundation models across diverse domains and tasks, revealing limitations in current models and introducing DA-Next-5M and DA-Next to advance spatial representation learning.

While spatial foundation models have demonstrated impressive performance on standard datasets, a critical question remains: are they truly all-round players capable of generalizing robustly across diverse downstream tasks, arbitrary viewpoints, shifting scene domains, varying input densities, and specific hardware constraints? Answering this overarching question requires a holistic assessment, yet current models are mainly evaluated on specific domains for which they were specifically designed or trained. Such evaluations are intrinsically limited by narrow paradigm coverage, limited scene domains, and arbitrary frame sampling, making it fundamentally difficult to assess their true generalization capabilities. To address this gap, we present SpatialBench, a cross-paradigm, domain-diverse benchmark for spatial foundation models with deterministic sampling. SpatialBench features unprecedented scale and rigorous deterministic design, comprising 19 datasets and 546 scenes across 5 diverse spatial domains. It comprehensively evaluates 41 models across 6 paradigms on 5 task suites under 4 different input density settings. Our extensive evaluation reveals that current models are not yet all-round players, and uncovers crucial insights for future advancement. Specifically, we demonstrate that full-context attention maximizes accuracy while bounded-memory strategies unlock long-sequence scalability. Moreover, our empirical evaluations in challenging embodied and egocentric tasks demonstrate that strict domain alignment and high data quality are far more critical to performance than simple dataset scaling. Furthermore, to address the largest data gap identified in our analysis, we go beyond evaluation by introducing a large-scale dataset, DA-Next-5M, and a strong baseline model, DA-Next, pushing the boundaries of spatial representation learning.
Original Article
View Cached Full Text

Cached at: 05/27/26, 02:47 AM

Paper page - SpatialBench: Is Your Spatial Foundation Model an All-Round Player?

Source: https://huggingface.co/papers/2605.27367 Published on May 26

·

Submitted byhttps://huggingface.co/lifuguan

leolion May 27

Authors:

,

,

,

,

,

,

,

,

,

,

,

Abstract

SpatialBench presents a comprehensive benchmark for evaluating spatial foundation models across diverse domains and tasks, revealing limitations in current models and introducing DA-Next-5M and DA-Next to advance spatial representation learning.

Whilespatial foundation modelshave demonstrated impressive performance on standard datasets, a critical question remains: are they truly all-round players capable of generalizing robustly across diverse downstream tasks, arbitrary viewpoints, shifting scene domains, varying input densities, and specific hardware constraints? Answering this overarching question requires a holistic assessment, yet current models are mainly evaluated on specific domains for which they were specifically designed or trained. Such evaluations are intrinsically limited by narrow paradigm coverage, limited scene domains, and arbitrary frame sampling, making it fundamentally difficult to assess their true generalization capabilities. To address this gap, we present SpatialBench, across-paradigm,domain-diverse benchmarkforspatial foundation modelswithdeterministic sampling. SpatialBench features unprecedented scale and rigorous deterministic design, comprising 19 datasets and 546 scenes across 5 diversespatial domains. It comprehensively evaluates 41 models across 6 paradigms on 5task suitesunder 4 differentinput density settings. Our extensive evaluation reveals that current models are not yet all-round players, and uncovers crucial insights for future advancement. Specifically, we demonstrate thatfull-context attentionmaximizes accuracy whilebounded-memory strategiesunlocklong-sequence scalability. Moreover, our empirical evaluations in challenging embodied andegocentric tasksdemonstrate that strict domain alignment and high data quality are far more critical to performance than simple dataset scaling. Furthermore, to address the largest data gap identified in our analysis, we go beyond evaluation by introducing a large-scale dataset, DA-Next-5M, and a strong baseline model, DA-Next, pushing the boundaries ofspatial representation learning.

View arXiv pageView PDFProject pageGitHub5Add to collection

Get this paper in your agent:

hf papers read 2605\.27367

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2605.27367 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2605.27367 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2605.27367 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

PlanBench-V: A Spatial Planning Map Benchmark for Vision-Language Models

arXiv cs.CL

This paper introduces PlanBench-V, the first comprehensive benchmark for evaluating Vision-Language Models on spatial planning map interpretation, including an expert-annotated dataset and a four-dimension evaluation framework. Experiments show significant progress but highlight persistent challenges in implementation-oriented tasks.

OVO-S-Bench: A Hierarchical Benchmark for Streaming Spatial Intelligence in Multimodal LLMs

Papers with Code Trending

OVO-S-Bench introduces a comprehensive human-annotated benchmark of 1,680 questions across 348 videos to evaluate streaming spatial intelligence in multimodal LLMs, revealing that even the best model (Gemini-3.1-Pro) trails human experts by 27 points. The benchmark exposes key limitations including allocentric mapping as a major bottleneck and chain-of-thought reasoning amplifying spatial errors.

Exploring Spatial Intelligence from a Generative Perspective

Hugging Face Daily Papers

Researchers introduce GSI-Bench, the first benchmark to quantify generative spatial intelligence in multimodal models by evaluating 3D spatial constraint compliance during image generation. Fine-tuning on their synthetic dataset boosts both spatial editing fidelity and downstream spatial understanding, showing generative training can strengthen spatial reasoning.