@inproceedings{li-etal-2026-webtraveler,
title = "{W}eb{T}raveler: A Multi-Agent Framework for Continuous Exploration in Website Traversal",
author = "Li, Rongchang and
Shen, Yang and
Tang, Gan and
Zhang, Zizhao and
Zhang, Shuai and
WU, Haipang and
Li, Xi",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.903/",
pages = "20800--20820",
ISBN = "979-8-89176-459-0",
abstract = "Large language models (LLMs) are increasingly used as web-facing agents for information-seeking tasks over live websites. Many such tasks require multi-step in-site navigation and cross-page evidence aggregation, a setting known as web traversal. Existing LLM-based web agents, including recent explore{--}critic architectures, typically store interaction history as an ever-growing unstructured text trace and mix high-level planning with low-level page actions. This makes long-horizon behavior brittle, obscures the source of evidence, and often leads to redundant browsing or premature stopping. To mitigate these issues, we propose WebTraveler, a hierarchical multi-agent framework that separates high-level information-seeking planning from local web exploration and maintains a structured trajectory memory of decision-relevant signals. This design provides compact information interfaces for evidence sufficiency checking and webpage navigation, improving long-horizon stability and cross-page evidence aggregation. To obtain scalable supervision, we further introduce path-guided data synthesis, which uses executable navigation paths to collect trajectories, removes explicit path hints, and filters invalid samples through answer-consistency checks. Experiments show that WebTraveler consistently outperforms previous methods across LLM scales, especially on tasks requiring deeper navigation, with further gains from path-guided supervision. Moreover, WebTraveler also shows generalization to broader web navigation tasks. Code is available at https://github.com/RongchangLi/WebTraveler."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-etal-2026-webtraveler">
<titleInfo>
<title>WebTraveler: A Multi-Agent Framework for Continuous Exploration in Website Traversal</title>
</titleInfo>
<name type="personal">
<namePart type="given">Rongchang</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yang</namePart>
<namePart type="family">Shen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gan</namePart>
<namePart type="family">Tang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zizhao</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shuai</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haipang</namePart>
<namePart type="family">WU</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xi</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Large language models (LLMs) are increasingly used as web-facing agents for information-seeking tasks over live websites. Many such tasks require multi-step in-site navigation and cross-page evidence aggregation, a setting known as web traversal. Existing LLM-based web agents, including recent explore–critic architectures, typically store interaction history as an ever-growing unstructured text trace and mix high-level planning with low-level page actions. This makes long-horizon behavior brittle, obscures the source of evidence, and often leads to redundant browsing or premature stopping. To mitigate these issues, we propose WebTraveler, a hierarchical multi-agent framework that separates high-level information-seeking planning from local web exploration and maintains a structured trajectory memory of decision-relevant signals. This design provides compact information interfaces for evidence sufficiency checking and webpage navigation, improving long-horizon stability and cross-page evidence aggregation. To obtain scalable supervision, we further introduce path-guided data synthesis, which uses executable navigation paths to collect trajectories, removes explicit path hints, and filters invalid samples through answer-consistency checks. Experiments show that WebTraveler consistently outperforms previous methods across LLM scales, especially on tasks requiring deeper navigation, with further gains from path-guided supervision. Moreover, WebTraveler also shows generalization to broader web navigation tasks. Code is available at https://github.com/RongchangLi/WebTraveler.</abstract>
<identifier type="citekey">li-etal-2026-webtraveler</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.903/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>20800</start>
<end>20820</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T WebTraveler: A Multi-Agent Framework for Continuous Exploration in Website Traversal
%A Li, Rongchang
%A Shen, Yang
%A Tang, Gan
%A Zhang, Zizhao
%A Zhang, Shuai
%A WU, Haipang
%A Li, Xi
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F li-etal-2026-webtraveler
%X Large language models (LLMs) are increasingly used as web-facing agents for information-seeking tasks over live websites. Many such tasks require multi-step in-site navigation and cross-page evidence aggregation, a setting known as web traversal. Existing LLM-based web agents, including recent explore–critic architectures, typically store interaction history as an ever-growing unstructured text trace and mix high-level planning with low-level page actions. This makes long-horizon behavior brittle, obscures the source of evidence, and often leads to redundant browsing or premature stopping. To mitigate these issues, we propose WebTraveler, a hierarchical multi-agent framework that separates high-level information-seeking planning from local web exploration and maintains a structured trajectory memory of decision-relevant signals. This design provides compact information interfaces for evidence sufficiency checking and webpage navigation, improving long-horizon stability and cross-page evidence aggregation. To obtain scalable supervision, we further introduce path-guided data synthesis, which uses executable navigation paths to collect trajectories, removes explicit path hints, and filters invalid samples through answer-consistency checks. Experiments show that WebTraveler consistently outperforms previous methods across LLM scales, especially on tasks requiring deeper navigation, with further gains from path-guided supervision. Moreover, WebTraveler also shows generalization to broader web navigation tasks. Code is available at https://github.com/RongchangLi/WebTraveler.
%U https://aclanthology.org/2026.emnlp-main.903/
%P 20800-20820
Markdown (Informal)
[WebTraveler: A Multi-Agent Framework for Continuous Exploration in Website Traversal](https://aclanthology.org/2026.emnlp-main.903/) (Li et al., EMNLP 2026)
ACL