SimuRA: A World-Model-Driven Simulative Reasoning Architecture for General Goal-Oriented Agents Article Swipe

PDF

Mingkai Deng , Jing Hou , Yilin Shen , Hongxia Jin , Graham Neubig , Zhiting Hu , Eric P. Xing ·

YOU? · · 2025 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2507.23773

AI agents built on foundation models hold enormous promise. Current practice, however, focuses on a one-task-one-agent approach, which not only falls short of scalability and generality, but also faces practical limitations from black-box autoregressive reasoning, where decisions unfold token by token without explicit simulation or counterfactual evaluation of outcomes. Humans, on the other hand, reason and plan by mentally simulating the consequences of actions within an internal model of the world -- a capability that supports flexible, goal-directed behavior across diverse contexts. Moving towards a more general and powerful AI agent, we introduce SimuRA, a goal-oriented architecture for generalized agentic reasoning. Based on a principled formulation of an optimal agent in any general environment, SimuRA addresses the limitations of black-box autoregressive reasoning by incorporating the world model for planning via simulation. Our prototype world model is implemented using LLMs as a substrate, leveraging the natural language as a discrete, hierarchical representation grounded in concepts for planning, while remaining model-agnostic. On complex web-browsing tasks such as flight search, SimuRA improves the success rate from 0% to 32.2% compared to a representative open-web agent baseline. Across tasks, world-model-based planning achieves up to 124% higher task completion rates than a matched black-box autoregressive baseline, demonstrating the advantages of simulative reasoning. We release ReasonerAgent-Web, a web-browsing agent built on SimuRA, as an open-source research demo.

Related Topics

Truth And Reconciliation Commission Of Canada

Reich Ministry Of Public Enlightenment And Propaganda

Rick Hurst

Mutual Aid

Fuck

Concepts

No concepts available.

Metadata

Type: preprint
Language: en
Landing Page: http://arxiv.org/abs/2507.23773
PDF: https://arxiv.org/pdf/2507.23773
OA Status: green
OpenAlex ID: https://openalex.org/W4414922302

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4414922302

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.48550/arxiv.2507.23773

Digital Object Identifier
Title: SimuRA: A World-Model-Driven Simulative Reasoning Architecture for General Goal-Oriented Agents

Work title
Type: preprint

OpenAlex work type
Language: en

Primary language
Publication year: 2025

Year of publication
Publication date: 2025-07-31

Full publication date if available
Authors: Mingkai Deng, Jing Hou, Yilin Shen, Hongxia Jin, Graham Neubig, Zhiting Hu, Eric P. Xing

List of authors in order
Landing page: https://arxiv.org/abs/2507.23773

Publisher landing page
PDF URL: https://arxiv.org/pdf/2507.23773

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/2507.23773

Direct OA link when available
Cited by: 0

Total citation count in OpenAlex

Full payload

id	https://openalex.org/W4414922302
doi	https://doi.org/10.48550/arxiv.2507.23773
ids.doi	https://doi.org/10.48550/arxiv.2507.23773
ids.openalex	https://openalex.org/W4414922302
fwci
type	preprint
title	SimuRA: A World-Model-Driven Simulative Reasoning Architecture for General Goal-Oriented Agents
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
topics[0].id	https://openalex.org/T10456
topics[0].field.id	https://openalex.org/fields/17
topics[0].field.display_name	Computer Science
topics[0].score	0.9743000268936157
topics[0].domain.id	https://openalex.org/domains/3
topics[0].domain.display_name	Physical Sciences
topics[0].subfield.id	https://openalex.org/subfields/1702
topics[0].subfield.display_name	Artificial Intelligence
topics[0].display_name	Multi-Agent Systems and Negotiation
topics[1].id	https://openalex.org/T10215
topics[1].field.id	https://openalex.org/fields/17
topics[1].field.display_name	Computer Science
topics[1].score	0.9699000120162964
topics[1].domain.id	https://openalex.org/domains/3
topics[1].domain.display_name	Physical Sciences
topics[1].subfield.id	https://openalex.org/subfields/1702
topics[1].subfield.display_name	Artificial Intelligence
topics[1].display_name	Semantic Web and Ontologies
topics[2].id	https://openalex.org/T10906
topics[2].field.id	https://openalex.org/fields/17
topics[2].field.display_name	Computer Science
topics[2].score	0.919700026512146
topics[2].domain.id	https://openalex.org/domains/3
topics[2].domain.display_name	Physical Sciences
topics[2].subfield.id	https://openalex.org/subfields/1702
topics[2].subfield.display_name	Artificial Intelligence
topics[2].display_name	AI-based Problem Solving and Planning
is_xpac	False
apc_list
apc_paid
language	en
locations[0].id	pmh:oai:arXiv.org:2507.23773
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/2507.23773
locations[0].version	submittedVersion
locations[0].raw_type
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/2507.23773
locations[1].id	doi:10.48550/arxiv.2507.23773
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
locations[1].pdf_url
locations[1].version
locations[1].raw_type	article
locations[1].license_id
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.2507.23773
indexed_in	arxiv, datacite
authorships[0].author.id	https://openalex.org/A5006031253
authorships[0].author.orcid
authorships[0].author.display_name	Mingkai Deng
authorships[0].author_position	first
authorships[0].raw_author_name	Deng, Mingkai
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5101756256
authorships[1].author.orcid	https://orcid.org/0000-0003-0792-8971
authorships[1].author.display_name	Jing Hou
authorships[1].author_position	middle
authorships[1].raw_author_name	Hou, Jinyu
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5058974318
authorships[2].author.orcid	https://orcid.org/0000-0002-1955-1529
authorships[2].author.display_name	Yilin Shen
authorships[2].author_position	middle
authorships[2].raw_author_name	Shen, Yilin
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5044205212
authorships[3].author.orcid	https://orcid.org/0009-0000-0222-4217
authorships[3].author.display_name	Hongxia Jin
authorships[3].author_position	middle
authorships[3].raw_author_name	Jin, Hongxia
authorships[3].is_corresponding	False
authorships[4].author.id	https://openalex.org/A5068811427
authorships[4].author.orcid	https://orcid.org/0000-0002-2072-3789
authorships[4].author.display_name	Graham Neubig
authorships[4].author_position	middle
authorships[4].raw_author_name	Neubig, Graham
authorships[4].is_corresponding	False
authorships[5].author.id	https://openalex.org/A5114674037
authorships[5].author.orcid
authorships[5].author.display_name	Zhiting Hu
authorships[5].author_position	middle
authorships[5].raw_author_name	Hu, Zhiting
authorships[5].is_corresponding	False
authorships[6].author.id	https://openalex.org/A5009547049
authorships[6].author.orcid	https://orcid.org/0009-0005-9158-4201
authorships[6].author.display_name	Eric P. Xing
authorships[6].author_position	last
authorships[6].raw_author_name	Xing, Eric
authorships[6].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/2507.23773
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2025-10-08T00:00:00
display_name	SimuRA: A World-Model-Driven Simulative Reasoning Architecture for General Goal-Oriented Agents
has_fulltext	False
is_retracted	False
updated_date	2025-11-06T06:51:31.235846
primary_topic.id	https://openalex.org/T10456
primary_topic.field.id	https://openalex.org/fields/17
primary_topic.field.display_name	Computer Science
primary_topic.score	0.9743000268936157
primary_topic.domain.id	https://openalex.org/domains/3
primary_topic.domain.display_name	Physical Sciences
primary_topic.subfield.id	https://openalex.org/subfields/1702
primary_topic.subfield.display_name	Artificial Intelligence
primary_topic.display_name	Multi-Agent Systems and Negotiation
cited_by_count	0
locations_count	2
best_oa_location.id	pmh:oai:arXiv.org:2507.23773
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/2507.23773
best_oa_location.version	submittedVersion
best_oa_location.raw_type
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/2507.23773
primary_location.id	pmh:oai:arXiv.org:2507.23773
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/2507.23773
primary_location.version	submittedVersion
primary_location.raw_type
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/2507.23773
publication_date	2025-07-31
publication_year	2025
referenced_works_count	0
abstract_inverted_index.a	14, 72, 84, 94, 103, 140, 147, 178, 196, 210
abstract_inverted_index.--	71
abstract_inverted_index.0%	173
abstract_inverted_index.AI	0, 89
abstract_inverted_index.On	159
abstract_inverted_index.We	207
abstract_inverted_index.an	65, 107, 217
abstract_inverted_index.as	139, 146, 164, 216
abstract_inverted_index.by	39, 57, 122
abstract_inverted_index.in	110, 152
abstract_inverted_index.is	135
abstract_inverted_index.of	22, 47, 62, 68, 106, 118, 204
abstract_inverted_index.on	3, 13, 50, 102, 214
abstract_inverted_index.or	44
abstract_inverted_index.to	174, 177, 189
abstract_inverted_index.up	188
abstract_inverted_index.we	91
abstract_inverted_index.Our	131
abstract_inverted_index.and	24, 55, 87
abstract_inverted_index.any	111
abstract_inverted_index.but	26
abstract_inverted_index.for	97, 127, 154
abstract_inverted_index.not	18
abstract_inverted_index.the	51, 60, 69, 116, 124, 143, 169, 202
abstract_inverted_index.via	129
abstract_inverted_index.124%	190
abstract_inverted_index.LLMs	138
abstract_inverted_index.also	27
abstract_inverted_index.from	31, 172
abstract_inverted_index.hold	6
abstract_inverted_index.more	85
abstract_inverted_index.only	19
abstract_inverted_index.plan	56
abstract_inverted_index.rate	171
abstract_inverted_index.such	163
abstract_inverted_index.task	192
abstract_inverted_index.than	195
abstract_inverted_index.that	74
abstract_inverted_index.32.2%	175
abstract_inverted_index.Based	101
abstract_inverted_index.agent	109, 181, 212
abstract_inverted_index.built	2, 213
abstract_inverted_index.demo.	220
abstract_inverted_index.faces	28
abstract_inverted_index.falls	20
abstract_inverted_index.hand,	53
abstract_inverted_index.model	67, 126, 134
abstract_inverted_index.other	52
abstract_inverted_index.rates	194
abstract_inverted_index.short	21
abstract_inverted_index.tasks	162
abstract_inverted_index.token	38, 40
abstract_inverted_index.using	137
abstract_inverted_index.where	35
abstract_inverted_index.which	17
abstract_inverted_index.while	156
abstract_inverted_index.world	70, 125, 133
abstract_inverted_index.Across	183
abstract_inverted_index.Moving	82
abstract_inverted_index.SimuRA	114, 167
abstract_inverted_index.across	79
abstract_inverted_index.agent,	90
abstract_inverted_index.agents	1
abstract_inverted_index.flight	165
abstract_inverted_index.higher	191
abstract_inverted_index.models	5
abstract_inverted_index.reason	54
abstract_inverted_index.tasks,	184
abstract_inverted_index.unfold	37
abstract_inverted_index.within	64
abstract_inverted_index.Current	9
abstract_inverted_index.Humans,	49
abstract_inverted_index.SimuRA,	93, 215
abstract_inverted_index.actions	63
abstract_inverted_index.agentic	99
abstract_inverted_index.complex	160
abstract_inverted_index.diverse	80
abstract_inverted_index.focuses	12
abstract_inverted_index.general	86, 112
abstract_inverted_index.matched	197
abstract_inverted_index.natural	144
abstract_inverted_index.optimal	108
abstract_inverted_index.release	208
abstract_inverted_index.search,	166
abstract_inverted_index.success	170
abstract_inverted_index.towards	83
abstract_inverted_index.without	41
abstract_inverted_index.achieves	187
abstract_inverted_index.behavior	78
abstract_inverted_index.compared	176
abstract_inverted_index.concepts	153
abstract_inverted_index.enormous	7
abstract_inverted_index.explicit	42
abstract_inverted_index.grounded	151
abstract_inverted_index.however,	11
abstract_inverted_index.improves	168
abstract_inverted_index.internal	66
abstract_inverted_index.language	145
abstract_inverted_index.mentally	58
abstract_inverted_index.open-web	180
abstract_inverted_index.planning	128, 186
abstract_inverted_index.powerful	88
abstract_inverted_index.promise.	8
abstract_inverted_index.research	219
abstract_inverted_index.supports	75
abstract_inverted_index.addresses	115
abstract_inverted_index.approach,	16
abstract_inverted_index.baseline,	200
abstract_inverted_index.baseline.	182
abstract_inverted_index.black-box	32, 119, 198
abstract_inverted_index.contexts.	81
abstract_inverted_index.decisions	36
abstract_inverted_index.discrete,	148
abstract_inverted_index.flexible,	76
abstract_inverted_index.introduce	92
abstract_inverted_index.outcomes.	48
abstract_inverted_index.planning,	155
abstract_inverted_index.practical	29
abstract_inverted_index.practice,	10
abstract_inverted_index.prototype	132
abstract_inverted_index.reasoning	121
abstract_inverted_index.remaining	157
abstract_inverted_index.advantages	203
abstract_inverted_index.capability	73
abstract_inverted_index.completion	193
abstract_inverted_index.evaluation	46
abstract_inverted_index.foundation	4
abstract_inverted_index.leveraging	142
abstract_inverted_index.principled	104
abstract_inverted_index.reasoning,	34
abstract_inverted_index.reasoning.	100, 206
abstract_inverted_index.simulating	59
abstract_inverted_index.simulation	43
abstract_inverted_index.simulative	205
abstract_inverted_index.substrate,	141
abstract_inverted_index.formulation	105
abstract_inverted_index.generality,	25
abstract_inverted_index.generalized	98
abstract_inverted_index.implemented	136
abstract_inverted_index.limitations	30, 117
abstract_inverted_index.open-source	218
abstract_inverted_index.scalability	23
abstract_inverted_index.simulation.	130
abstract_inverted_index.architecture	96
abstract_inverted_index.consequences	61
abstract_inverted_index.environment,	113
abstract_inverted_index.hierarchical	149
abstract_inverted_index.web-browsing	161, 211
abstract_inverted_index.demonstrating	201
abstract_inverted_index.goal-directed	77
abstract_inverted_index.goal-oriented	95
abstract_inverted_index.incorporating	123
abstract_inverted_index.autoregressive	33, 120, 199
abstract_inverted_index.counterfactual	45
abstract_inverted_index.representation	150
abstract_inverted_index.representative	179
abstract_inverted_index.model-agnostic.	158
abstract_inverted_index.world-model-based	185
abstract_inverted_index.ReasonerAgent-Web,	209
abstract_inverted_index.one-task-one-agent	15
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	7
citation_normalized_percentile