SpatialBench：您的空间基础模型是全能选手吗？

摘要

尽管空间基础模型在标准数据集上展现了令人瞩目的性能，但一个关键问题仍然存在：它们是否真正具备全能型能力——能够在多样化下游任务、任意视角、不断变化的场景域、变化的输入密度以及特定硬件约束下实现稳健泛化？回答这一总体性问题需要全面评估，然而当前模型主要在其专门设计或训练过的特定领域进行评估。此类评估本质上受限于狭窄的范式覆盖范围、有限的场景域和任意帧采样，使得评估其真实泛化能力面临根本性困难。为填补这一空白，我们提出SpatialBench——一个面向空间基础模型的跨范式、多领域基准测试，采用确定性采样机制。SpatialBench具备前所未有的规模和严谨的确定性设计，涵盖5个不同空间领域的19个数据集和546个场景。它系统评估了6种范式下的41个模型在5个任务套件及4种不同输入密度设置下的表现。我们的广泛评估表明，当前模型尚未成为全能型选手，并揭示了未来发展的关键洞察：具体而言，全上下文注意力机制最大化精度，而有限内存策略解锁长序列可扩展性。此外，针对挑战性具身与自我中心任务的实证评估显示，严格的领域对齐和高质量数据对性能的影响远大于简单的数据集规模扩展。为填补分析中发现的最大数据缺口，我们超越评估范畴，引入大规模数据集DA-Next-5M和强基线模型DA-Next，进一步拓展空间表征学习的边界。

English

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.