@inproceedings{pang-etal-2025-assimilation,
title = "Assimilation and Accommodation: Task-Adaptive Hierarchical Abstraction for Solving Web Tasks",
author = "Pang, Xinyu and
Hong, Ruixin and
Zhang, Hongming and
Zhang, Changshui",
editor = "Che, Wanxiang and
Nabende, Joyce and
Shutova, Ekaterina and
Pilehvar, Mohammad Taher",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2025",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-acl.720/",
doi = "10.18653/v1/2025.findings-acl.720",
pages = "14000--14014",
ISBN = "979-8-89176-256-5",
abstract = "Web tasks, which involve processing data from online resources, challenge agents to generalize beyond fixed knowledge to unseen task contexts. Learning from experience, the ability to derive reusable patterns from past tasks, is crucial for improving generalization. However, existing methods focus on summarizing workflows, i.e., common sub-routines, which may introduce excessive low-level details that distract models. Additionally, the absence of task-specific objectives can lead to inconsistencies between workflows and future task queries, hindering reasoning performance. This paper seeks to mitigate these issues by proposing $A^2$, a framework that derives task-adaptive hierarchical abstraction to enhance web task reasoning. Our approach first extracts general-purpose semantic abstraction from past task-solution pairs. Combined with the next task query, this abstraction forms a task-adaptive episodic abstraction that guides subsequent reasoning. Experiments show that $A^2$ achieves superior performance with competitive cost-efficiency, improving success rates by 0.7{\%} on Mind2web and 4.6{\%} on Webarena."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="pang-etal-2025-assimilation">
<titleInfo>
<title>Assimilation and Accommodation: Task-Adaptive Hierarchical Abstraction for Solving Web Tasks</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xinyu</namePart>
<namePart type="family">Pang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruixin</namePart>
<namePart type="family">Hong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hongming</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Changshui</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wanxiang</namePart>
<namePart type="family">Che</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joyce</namePart>
<namePart type="family">Nabende</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Shutova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="given">Taher</namePart>
<namePart type="family">Pilehvar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Vienna, Austria</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-256-5</identifier>
</relatedItem>
<abstract>Web tasks, which involve processing data from online resources, challenge agents to generalize beyond fixed knowledge to unseen task contexts. Learning from experience, the ability to derive reusable patterns from past tasks, is crucial for improving generalization. However, existing methods focus on summarizing workflows, i.e., common sub-routines, which may introduce excessive low-level details that distract models. Additionally, the absence of task-specific objectives can lead to inconsistencies between workflows and future task queries, hindering reasoning performance. This paper seeks to mitigate these issues by proposing A², a framework that derives task-adaptive hierarchical abstraction to enhance web task reasoning. Our approach first extracts general-purpose semantic abstraction from past task-solution pairs. Combined with the next task query, this abstraction forms a task-adaptive episodic abstraction that guides subsequent reasoning. Experiments show that A² achieves superior performance with competitive cost-efficiency, improving success rates by 0.7% on Mind2web and 4.6% on Webarena.</abstract>
<identifier type="citekey">pang-etal-2025-assimilation</identifier>
<identifier type="doi">10.18653/v1/2025.findings-acl.720</identifier>
<location>
<url>https://aclanthology.org/2025.findings-acl.720/</url>
</location>
<part>
<date>2025-07</date>
<extent unit="page">
<start>14000</start>
<end>14014</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Assimilation and Accommodation: Task-Adaptive Hierarchical Abstraction for Solving Web Tasks
%A Pang, Xinyu
%A Hong, Ruixin
%A Zhang, Hongming
%A Zhang, Changshui
%Y Che, Wanxiang
%Y Nabende, Joyce
%Y Shutova, Ekaterina
%Y Pilehvar, Mohammad Taher
%S Findings of the Association for Computational Linguistics: ACL 2025
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-256-5
%F pang-etal-2025-assimilation
%X Web tasks, which involve processing data from online resources, challenge agents to generalize beyond fixed knowledge to unseen task contexts. Learning from experience, the ability to derive reusable patterns from past tasks, is crucial for improving generalization. However, existing methods focus on summarizing workflows, i.e., common sub-routines, which may introduce excessive low-level details that distract models. Additionally, the absence of task-specific objectives can lead to inconsistencies between workflows and future task queries, hindering reasoning performance. This paper seeks to mitigate these issues by proposing A², a framework that derives task-adaptive hierarchical abstraction to enhance web task reasoning. Our approach first extracts general-purpose semantic abstraction from past task-solution pairs. Combined with the next task query, this abstraction forms a task-adaptive episodic abstraction that guides subsequent reasoning. Experiments show that A² achieves superior performance with competitive cost-efficiency, improving success rates by 0.7% on Mind2web and 4.6% on Webarena.
%R 10.18653/v1/2025.findings-acl.720
%U https://aclanthology.org/2025.findings-acl.720/
%U https://doi.org/10.18653/v1/2025.findings-acl.720
%P 14000-14014
Markdown (Informal)
[Assimilation and Accommodation: Task-Adaptive Hierarchical Abstraction for Solving Web Tasks](https://aclanthology.org/2025.findings-acl.720/) (Pang et al., Findings 2025)
ACL