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

• URL: http://arxiv.org/abs/2605.27367v2
• Date: Fri, 29 May 2026 09:30:04 GMT
• Title: SpatialBench: Is Your Spatial Foundation Model an All-Round Player?
• Abstract summary: SpatialBench is a cross-paradigm, domain-diverse benchmark for spatial foundation models with deterministic sampling.<n>It comprehensively evaluates 41 models across 6 paradigms on 5 task suites under 4 different input density settings.<n>Our 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.
• Score: 92.031716744172
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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.

Related papers

• AnyGroundBench: A Specialized-Domain Benchmark for Video Grounding in Vision-Language Models [15.610649996654876]
  We introduce AnyGroundBench, a benchmark designed to shift the STVG evaluation from static zero-shot testing to rigorous domain adaptation.<n>We target five specialized domains (animal, industry, sports, surgery, public security)<n>We extensively evaluate 15 state-of-the-art VLMs, assessing their zero-shot generalization and In-Context Learning (ICL) capabilities.
  arXiv  Detail & Related papers  (2026-07-02T14:52:45Z)
• DepthCharge: A Domain-Agnostic Framework for Measuring Depth-Dependent Knowledge in Large Language Models [51.56484100374058]
  Large Language Models appear competent when answering general questions but often fail when pushed into domain-specific details.<n>We present DepthCharge, a domain-agnostic framework that measures knowledge depth through three innovations.<n> adaptive probing that generates follow-up questions based on concepts the model actually mentions, on-demand fact verification from authoritative sources, and survival statistics with constant sample sizes at every depth level.
  arXiv  Detail & Related papers  (2026-03-05T20:49:11Z)
• It's TIME: Towards the Next Generation of Time Series Forecasting Benchmarks [87.7937890373758]
  Time series foundation models (TSFMs) are revolutionizing the forecasting landscape from specific dataset modeling to generalizable task evaluation.<n>We introduce TIME, a next-generation task-centric benchmark comprising 50 fresh datasets and 98 forecasting tasks.<n>We propose a novel pattern-level evaluation perspective that moves beyond traditional dataset-level evaluations based on static meta labels.
  arXiv  Detail & Related papers  (2026-02-12T16:31:01Z)
• TSRBench: A Comprehensive Multi-task Multi-modal Time Series Reasoning Benchmark for Generalist Models [52.21298691935726]
  The ability to reason over time series is a fundamental skill for generalist models to solve practical problems.<n>To bridge this gap, we introduce TSRBench, a comprehensive benchmark designed to stress-test the full spectrum of time series reasoning capabilities.
  arXiv  Detail & Related papers  (2026-01-26T18:04:54Z)
• Your One-Stop Solution for AI-Generated Video Detection [26.581301251283943]
  generative modeling can create remarkably realistic synthetic videos.<n>However, two key limitations hinder the development of this field.<n>We propose AIGVDBench, a benchmark designed to be comprehensive and representative.
  arXiv  Detail & Related papers  (2026-01-16T07:02:06Z)
• ROVR-Open-Dataset: A Large-Scale Depth Dataset for Autonomous Driving [62.9051914830949]
  We present ROVR, a large-scale, diverse, and cost-efficient depth dataset designed to capture the complexity of real-world driving.<n>A lightweight acquisition pipeline ensures scalable collection, while sparse but statistically sufficient ground truth supports robust training.<n> Benchmarking with state-of-the-art monocular depth models reveals severe cross-dataset generalization failures.
  arXiv  Detail & Related papers  (2025-08-19T16:13:49Z)
• Vision-G1: Towards General Vision Language Reasoning with Multi-Domain Data Curation [64.23194519770897]
  We build a comprehensive RL-ready visual reasoning dataset from 46 data sources across 8 dimensions.<n>We propose an influence function based data selection and difficulty based filtering strategy to identify high-quality training samples from this dataset.<n>We train the VLM, referred to as Vision-G1, using multi-round RL with a data curriculum to iteratively improve its visual reasoning capabilities.
  arXiv  Detail & Related papers  (2025-08-18T07:24:33Z)
• ViewSpatial-Bench: Evaluating Multi-perspective Spatial Localization in Vision-Language Models [68.46716645478661]
  Vision-language models (VLMs) have demonstrated remarkable capabilities in understanding and reasoning about visual content.<n>Current VLMs excel primarily at egocentric spatial reasoning (from the camera's perspective) but fail to generalize to allocentric viewpoints.<n>We introduce ViewSpatial-Bench, the first comprehensive benchmark designed specifically for multi-viewpoint spatial localization recognition evaluation.
  arXiv  Detail & Related papers  (2025-05-27T17:59:26Z)
• Spatial457: A Diagnostic Benchmark for 6D Spatial Reasoning of Large Multimodal Models [8.499125564147834]
  We present a scalable and unbiased synthetic dataset designed with 4 key capability for spatial reasoning.<n>We develop a cascading evaluation structure, constructing 7 question types across 5 difficulty levels.<n>We observe a general decline in performance as task increases complexity, particularly in 3D reasoning and 6D spatial tasks.
  arXiv  Detail & Related papers  (2025-02-12T18:53:20Z)
• MSTAR: Multi-Scale Backbone Architecture Search for Timeseries Classification [0.41185655356953593]
  We propose a novel multi-scale search space and a framework for Neural architecture search (NAS) We show that our model can serve as a backbone to employ a powerful Transformer module with both untrained and pre-trained weights. Our search space reaches the state-of-the-art performance on four datasets on four different domains.
  arXiv  Detail & Related papers  (2024-02-21T13:59:55Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.