DSI-Bench: A Benchmark for Dynamic Spatial Intelligence Article Swipe

PDF

Ziang Zhang , Z. Jane Wang , Guanghao Zhang , Weilong Dai , Ying Xia , Zhennan Yan , Minjie Hong , Zhou Zhao ·

YOU? · · 2025 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2510.18873

Reasoning about dynamic spatial relationships is essential, as both observers and objects often move simultaneously. Although vision-language models (VLMs) and visual expertise models excel in 2D tasks and static scenarios, their ability to fully understand dynamic 3D scenarios remains limited. We introduce Dynamic Spatial Intelligence and propose DSI-Bench, a benchmark with nearly 1,000 dynamic videos and over 1,700 manually annotated questions covering nine decoupled motion patterns of observers and objects. Spatially and temporally symmetric designs reduce biases and enable systematic evaluation of models' reasoning about self-motion and object motion. Our evaluation of 14 VLMs and expert models reveals key limitations: models often conflate observer and object motion, exhibit semantic biases, and fail to accurately infer relative relationships in dynamic scenarios. Our DSI-Bench provides valuable findings and insights about the future development of general and expertise models with dynamic spatial intelligence.

Related Topics

Dynamic Programming

Dynamic Random-Access Memory

Dynamic Time Warping

Benchmark (Computing)

Dynamic Currency Conversion

Dynamic Web Page

Dynamic Positioning

Benchmark (Surveying)

Synchronous Dynamic Random-Access Memory

Dynamic Html

Dynamic Systems Development Method

Concepts

No concepts available.

Metadata

Type: preprint
Landing Page: http://arxiv.org/abs/2510.18873
PDF: https://arxiv.org/pdf/2510.18873
OA Status: green
OpenAlex ID: https://openalex.org/W4416057657

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4416057657

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.48550/arxiv.2510.18873

Digital Object Identifier
Title: DSI-Bench: A Benchmark for Dynamic Spatial Intelligence

Work title
Type: preprint

OpenAlex work type
Publication year: 2025

Year of publication
Publication date: 2025-10-21

Full publication date if available
Authors: Ziang Zhang, Z. Jane Wang, Guanghao Zhang, Weilong Dai, Ying Xia, Zhennan Yan, Minjie Hong, Zhou Zhao

List of authors in order
Landing page: https://arxiv.org/abs/2510.18873

Publisher landing page
PDF URL: https://arxiv.org/pdf/2510.18873

Direct link to full text PDF
Open access: Yes

Whether a free full text is available
OA status: green

Open access status per OpenAlex
OA URL: https://arxiv.org/pdf/2510.18873

Direct OA link when available
Cited by: 0

Total citation count in OpenAlex

Full payload

id	https://openalex.org/W4416057657
doi	https://doi.org/10.48550/arxiv.2510.18873
ids.doi	https://doi.org/10.48550/arxiv.2510.18873
ids.openalex	https://openalex.org/W4416057657
fwci
type	preprint
title	DSI-Bench: A Benchmark for Dynamic Spatial Intelligence
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
is_xpac	False
apc_list
apc_paid
language
locations[0].id	pmh:oai:arXiv.org:2510.18873
locations[0].is_oa	True
locations[0].source.id	https://openalex.org/S4306400194
locations[0].source.issn
locations[0].source.type	repository
locations[0].source.is_oa	True
locations[0].source.issn_l
locations[0].source.is_core	False
locations[0].source.is_in_doaj	False
locations[0].source.display_name	arXiv (Cornell University)
locations[0].source.host_organization	https://openalex.org/I205783295
locations[0].source.host_organization_name	Cornell University
locations[0].source.host_organization_lineage	https://openalex.org/I205783295
locations[0].license
locations[0].pdf_url	https://arxiv.org/pdf/2510.18873
locations[0].version	submittedVersion
locations[0].raw_type	text
locations[0].license_id
locations[0].is_accepted	False
locations[0].is_published	False
locations[0].raw_source_name
locations[0].landing_page_url	http://arxiv.org/abs/2510.18873
locations[1].id	doi:10.48550/arxiv.2510.18873
locations[1].is_oa	True
locations[1].source.id	https://openalex.org/S4306400194
locations[1].source.issn
locations[1].source.type	repository
locations[1].source.is_oa	True
locations[1].source.issn_l
locations[1].source.is_core	False
locations[1].source.is_in_doaj	False
locations[1].source.display_name	arXiv (Cornell University)
locations[1].source.host_organization	https://openalex.org/I205783295
locations[1].source.host_organization_name	Cornell University
locations[1].source.host_organization_lineage	https://openalex.org/I205783295
locations[1].license	cc-by
locations[1].pdf_url
locations[1].version
locations[1].raw_type	article
locations[1].license_id	https://openalex.org/licenses/cc-by
locations[1].is_accepted	False
locations[1].is_published
locations[1].raw_source_name
locations[1].landing_page_url	https://doi.org/10.48550/arxiv.2510.18873
indexed_in	arxiv, datacite
authorships[0].author.id	https://openalex.org/A5101538149
authorships[0].author.orcid	https://orcid.org/0000-0002-4551-3489
authorships[0].author.display_name	Ziang Zhang
authorships[0].author_position	first
authorships[0].raw_author_name	Zhang, Ziang
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5051587486
authorships[1].author.orcid	https://orcid.org/0000-0002-3791-0249
authorships[1].author.display_name	Z. Jane Wang
authorships[1].author_position	middle
authorships[1].raw_author_name	Wang, Zehan
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5101450274
authorships[2].author.orcid	https://orcid.org/0000-0002-6445-2271
authorships[2].author.display_name	Guanghao Zhang
authorships[2].author_position	middle
authorships[2].raw_author_name	Zhang, Guanghao
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5102266633
authorships[3].author.orcid	https://orcid.org/0009-0006-9288-932X
authorships[3].author.display_name	Weilong Dai
authorships[3].author_position	middle
authorships[3].raw_author_name	Dai, Weilong
authorships[3].is_corresponding	False
authorships[4].author.id	https://openalex.org/A5029615383
authorships[4].author.orcid	https://orcid.org/0000-0003-4415-4100
authorships[4].author.display_name	Ying Xia
authorships[4].author_position	middle
authorships[4].raw_author_name	Xia, Yan
authorships[4].is_corresponding	False
authorships[5].author.id	https://openalex.org/A5077384503
authorships[5].author.orcid	https://orcid.org/0000-0001-7128-1696
authorships[5].author.display_name	Zhennan Yan
authorships[5].author_position	middle
authorships[5].raw_author_name	Yan, Ziang
authorships[5].is_corresponding	False
authorships[6].author.id	https://openalex.org/A5114213524
authorships[6].author.orcid
authorships[6].author.display_name	Minjie Hong
authorships[6].author_position	middle
authorships[6].raw_author_name	Hong, Minjie
authorships[6].is_corresponding	False
authorships[7].author.id	https://openalex.org/A5079260216
authorships[7].author.orcid	https://orcid.org/0000-0001-6121-0384
authorships[7].author.display_name	Zhou Zhao
authorships[7].author_position	last
authorships[7].raw_author_name	Zhao, Zhou
authorships[7].is_corresponding	False
has_content.pdf	False
has_content.grobid_xml	False
is_paratext	False
open_access.is_oa	True
open_access.oa_url	https://arxiv.org/pdf/2510.18873
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2025-10-24T00:00:00
display_name	DSI-Bench: A Benchmark for Dynamic Spatial Intelligence
has_fulltext	False
is_retracted	False
updated_date	2025-11-28T10:36:26.776369
primary_topic
cited_by_count	0
locations_count	2
best_oa_location.id	pmh:oai:arXiv.org:2510.18873
best_oa_location.is_oa	True
best_oa_location.source.id	https://openalex.org/S4306400194
best_oa_location.source.issn
best_oa_location.source.type	repository
best_oa_location.source.is_oa	True
best_oa_location.source.issn_l
best_oa_location.source.is_core	False
best_oa_location.source.is_in_doaj	False
best_oa_location.source.display_name	arXiv (Cornell University)
best_oa_location.source.host_organization	https://openalex.org/I205783295
best_oa_location.source.host_organization_name	Cornell University
best_oa_location.source.host_organization_lineage	https://openalex.org/I205783295
best_oa_location.license
best_oa_location.pdf_url	https://arxiv.org/pdf/2510.18873
best_oa_location.version	submittedVersion
best_oa_location.raw_type	text
best_oa_location.license_id
best_oa_location.is_accepted	False
best_oa_location.is_published	False
best_oa_location.raw_source_name
best_oa_location.landing_page_url	http://arxiv.org/abs/2510.18873
primary_location.id	pmh:oai:arXiv.org:2510.18873
primary_location.is_oa	True
primary_location.source.id	https://openalex.org/S4306400194
primary_location.source.issn
primary_location.source.type	repository
primary_location.source.is_oa	True
primary_location.source.issn_l
primary_location.source.is_core	False
primary_location.source.is_in_doaj	False
primary_location.source.display_name	arXiv (Cornell University)
primary_location.source.host_organization	https://openalex.org/I205783295
primary_location.source.host_organization_name	Cornell University
primary_location.source.host_organization_lineage	https://openalex.org/I205783295
primary_location.license
primary_location.pdf_url	https://arxiv.org/pdf/2510.18873
primary_location.version	submittedVersion
primary_location.raw_type	text
primary_location.license_id
primary_location.is_accepted	False
primary_location.is_published	False
primary_location.raw_source_name
primary_location.landing_page_url	http://arxiv.org/abs/2510.18873
publication_date	2025-10-21
publication_year	2025
referenced_works_count	0
abstract_inverted_index.a	48
abstract_inverted_index.14	92
abstract_inverted_index.2D	25
abstract_inverted_index.3D	36
abstract_inverted_index.We	40
abstract_inverted_index.as	7
abstract_inverted_index.in	24, 117
abstract_inverted_index.is	5
abstract_inverted_index.of	66, 81, 91, 131
abstract_inverted_index.to	32, 112
abstract_inverted_index.Our	89, 120
abstract_inverted_index.and	10, 19, 27, 45, 55, 68, 71, 77, 86, 94, 104, 110, 125, 133
abstract_inverted_index.key	98
abstract_inverted_index.the	128
abstract_inverted_index.VLMs	93
abstract_inverted_index.both	8
abstract_inverted_index.fail	111
abstract_inverted_index.move	13
abstract_inverted_index.nine	62
abstract_inverted_index.over	56
abstract_inverted_index.with	50, 136
abstract_inverted_index.1,000	52
abstract_inverted_index.1,700	57
abstract_inverted_index.about	1, 84, 127
abstract_inverted_index.excel	23
abstract_inverted_index.fully	33
abstract_inverted_index.infer	114
abstract_inverted_index.often	12, 101
abstract_inverted_index.tasks	26
abstract_inverted_index.their	30
abstract_inverted_index.(VLMs)	18
abstract_inverted_index.biases	76
abstract_inverted_index.enable	78
abstract_inverted_index.expert	95
abstract_inverted_index.future	129
abstract_inverted_index.models	17, 22, 96, 100, 135
abstract_inverted_index.motion	64
abstract_inverted_index.nearly	51
abstract_inverted_index.object	87, 105
abstract_inverted_index.reduce	75
abstract_inverted_index.static	28
abstract_inverted_index.videos	54
abstract_inverted_index.visual	20
abstract_inverted_index.Dynamic	42
abstract_inverted_index.Spatial	43
abstract_inverted_index.ability	31
abstract_inverted_index.biases,	109
abstract_inverted_index.designs	74
abstract_inverted_index.dynamic	2, 35, 53, 118, 137
abstract_inverted_index.exhibit	107
abstract_inverted_index.general	132
abstract_inverted_index.models'	82
abstract_inverted_index.motion,	106
abstract_inverted_index.motion.	88
abstract_inverted_index.objects	11
abstract_inverted_index.propose	46
abstract_inverted_index.remains	38
abstract_inverted_index.reveals	97
abstract_inverted_index.spatial	3, 138
abstract_inverted_index.Although	15
abstract_inverted_index.conflate	102
abstract_inverted_index.covering	61
abstract_inverted_index.findings	124
abstract_inverted_index.insights	126
abstract_inverted_index.limited.	39
abstract_inverted_index.manually	58
abstract_inverted_index.objects.	69
abstract_inverted_index.observer	103
abstract_inverted_index.patterns	65
abstract_inverted_index.provides	122
abstract_inverted_index.relative	115
abstract_inverted_index.semantic	108
abstract_inverted_index.valuable	123
abstract_inverted_index.DSI-Bench	121
abstract_inverted_index.Reasoning	0
abstract_inverted_index.Spatially	70
abstract_inverted_index.annotated	59
abstract_inverted_index.benchmark	49
abstract_inverted_index.decoupled	63
abstract_inverted_index.expertise	21, 134
abstract_inverted_index.introduce	41
abstract_inverted_index.observers	9, 67
abstract_inverted_index.questions	60
abstract_inverted_index.reasoning	83
abstract_inverted_index.scenarios	37
abstract_inverted_index.symmetric	73
abstract_inverted_index.DSI-Bench,	47
abstract_inverted_index.accurately	113
abstract_inverted_index.essential,	6
abstract_inverted_index.evaluation	80, 90
abstract_inverted_index.scenarios,	29
abstract_inverted_index.scenarios.	119
abstract_inverted_index.systematic	79
abstract_inverted_index.temporally	72
abstract_inverted_index.understand	34
abstract_inverted_index.development	130
abstract_inverted_index.self-motion	85
abstract_inverted_index.Intelligence	44
abstract_inverted_index.limitations:	99
abstract_inverted_index.intelligence.	139
abstract_inverted_index.relationships	4, 116
abstract_inverted_index.simultaneously.	14
abstract_inverted_index.vision-language	16
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	8
citation_normalized_percentile