@inproceedings{huang-etal-2026-webworld,
title = "{W}eb{W}orld: Web-Simulated World Model",
author = "Huang, Ting and
Zhang, Zeyu and
Tang, Hao",
editor = "Zhang, Yue and
Xu, Ruifeng and
Purwarianti, Ayu and
Rong, Tong and
Suzuki, Jun",
booktitle = "Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the {A}sia-Pacific Chapter of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = nov,
year = "2026",
address = "Hengqin, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.aacl-long.162/",
pages = "3592--3608",
ISBN = "979-8-89176-439-2",
abstract = "Large language models (LLMs) have recently shown promise as planners across web, spatial, and embodied domains. However, existing agents often handle information seeking, spatial grounding, and action planning as separate capabilities, limiting their ability to solve long-horizon compositional tasks. To address these challenges, we present **WebWorld**, a web-simulated world model for long-horizon reasoning over web evidence, scene observations, and action plans. WebWorld maintains structured intermediate states across heterogeneous modalities and verifies them through schema-guided hierarchical planning. The framework introduces a Mixture-of-Transformers architecture that coordinates website, scene, and motion experts via multi-modal self-attention. To support training and evaluation, we develop WebWorld-CoT-55K, a large-scale dataset that integrates heterogeneous modalities into schema-guided reasoning traces, and **WebWorldBench**, a curated benchmark for evaluating long-horizon compositional tasks. Training combines cold-start supervised fine-tuning with offline reinforcement learning based on structural, semantic, and planning rewards; inference uses test-time scaling to aggregate multiple reasoning trajectories. Experiments on WebWorldBench show that WebWorld improves the overall score by up to 8.4{\%} over strong LLM-agent baselines, with gains in outdoor reasoning, geolocation, and travel planning, while still leaving substantial room for progress toward human-level performance."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="huang-etal-2026-webworld">
<titleInfo>
<title>WebWorld: Web-Simulated World Model</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ting</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zeyu</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hao</namePart>
<namePart type="family">Tang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruifeng</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ayu</namePart>
<namePart type="family">Purwarianti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tong</namePart>
<namePart type="family">Rong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Suzuki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Hengqin, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-439-2</identifier>
</relatedItem>
<abstract>Large language models (LLMs) have recently shown promise as planners across web, spatial, and embodied domains. However, existing agents often handle information seeking, spatial grounding, and action planning as separate capabilities, limiting their ability to solve long-horizon compositional tasks. To address these challenges, we present **WebWorld**, a web-simulated world model for long-horizon reasoning over web evidence, scene observations, and action plans. WebWorld maintains structured intermediate states across heterogeneous modalities and verifies them through schema-guided hierarchical planning. The framework introduces a Mixture-of-Transformers architecture that coordinates website, scene, and motion experts via multi-modal self-attention. To support training and evaluation, we develop WebWorld-CoT-55K, a large-scale dataset that integrates heterogeneous modalities into schema-guided reasoning traces, and **WebWorldBench**, a curated benchmark for evaluating long-horizon compositional tasks. Training combines cold-start supervised fine-tuning with offline reinforcement learning based on structural, semantic, and planning rewards; inference uses test-time scaling to aggregate multiple reasoning trajectories. Experiments on WebWorldBench show that WebWorld improves the overall score by up to 8.4% over strong LLM-agent baselines, with gains in outdoor reasoning, geolocation, and travel planning, while still leaving substantial room for progress toward human-level performance.</abstract>
<identifier type="citekey">huang-etal-2026-webworld</identifier>
<location>
<url>https://aclanthology.org/2026.aacl-long.162/</url>
</location>
<part>
<date>2026-11</date>
<extent unit="page">
<start>3592</start>
<end>3608</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T WebWorld: Web-Simulated World Model
%A Huang, Ting
%A Zhang, Zeyu
%A Tang, Hao
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%S Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-439-2
%F huang-etal-2026-webworld
%X Large language models (LLMs) have recently shown promise as planners across web, spatial, and embodied domains. However, existing agents often handle information seeking, spatial grounding, and action planning as separate capabilities, limiting their ability to solve long-horizon compositional tasks. To address these challenges, we present **WebWorld**, a web-simulated world model for long-horizon reasoning over web evidence, scene observations, and action plans. WebWorld maintains structured intermediate states across heterogeneous modalities and verifies them through schema-guided hierarchical planning. The framework introduces a Mixture-of-Transformers architecture that coordinates website, scene, and motion experts via multi-modal self-attention. To support training and evaluation, we develop WebWorld-CoT-55K, a large-scale dataset that integrates heterogeneous modalities into schema-guided reasoning traces, and **WebWorldBench**, a curated benchmark for evaluating long-horizon compositional tasks. Training combines cold-start supervised fine-tuning with offline reinforcement learning based on structural, semantic, and planning rewards; inference uses test-time scaling to aggregate multiple reasoning trajectories. Experiments on WebWorldBench show that WebWorld improves the overall score by up to 8.4% over strong LLM-agent baselines, with gains in outdoor reasoning, geolocation, and travel planning, while still leaving substantial room for progress toward human-level performance.
%U https://aclanthology.org/2026.aacl-long.162/
%P 3592-3608
Markdown (Informal)
[WebWorld: Web-Simulated World Model](https://aclanthology.org/2026.aacl-long.162/) (Huang et al., AACL-IJCNLP 2026)
ACL
- Ting Huang, Zeyu Zhang, and Hao Tang. 2026. WebWorld: Web-Simulated World Model. In Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3592–3608, Hengqin, China. Association for Computational Linguistics.