<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="gabbasov-etal-2026-comparative">
    <titleInfo>
        <title>Comparative Study of AI Agent Architecture Efficiency in Table Processing Tasks</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Timur</namePart>
        <namePart type="family">Gabbasov</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Aleksandr</namePart>
        <namePart type="family">Sandalov</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Aleksandr</namePart>
        <namePart type="family">Maksimov</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Maria</namePart>
        <namePart type="family">Khodorchenko</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Nikolay</namePart>
        <namePart type="given">Alekseevich</namePart>
        <namePart type="family">Butakov</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>The Proceedings of Workshop on Document Intelligence and Understanding</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Santosh</namePart>
            <namePart type="family">T.Y.S.S.</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Amit</namePart>
            <namePart type="family">Agarwal</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Vivek</namePart>
            <namePart type="family">Gupta</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Yova</namePart>
            <namePart type="family">Kementchedjhieva</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Aparna</namePart>
            <namePart type="family">Garimella</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Yufang</namePart>
            <namePart type="family">Hou</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">979-8-89176-477-4</identifier>
    </relatedItem>
    <abstract>Tables are central to business document workflows, but practical table-processing tasks vary substantially in structure, content, and required action. We compare eight ReAct-agent configurations built on the same Qwen3.6-35B-A3B model. The design crosses four tool profiles - Minimal, Search, Spreadsheet, and Full - with the presence or absence of Markdown-based procedural skills. We evaluate the configurations on 40 tasks from company workflows, FinQA, and SpreadsheetBench. Across three complete replicates, comprising 960 agent-task episodes, Full with skills obtains the highest aggregate success rate of 75.0% and a mean score of 0.8061. Full without skills reaches 65.0%, while Spreadsheet without skills reaches 73.3%. Aggregated across tool profiles, skills increase success from 69.4% to 72.3%, but their effect is heterogeneous: negative for Minimal and Spreadsheet and positive for Search and Full. Overall success remains stable across replicates at 70.3%, 70.9%, and 71.3%, although individual configuration results vary more substantially. These findings suggest that procedural guidance is most promising when an agent must coordinate a broad tool environment rather than as a universal augmentation.</abstract>
    <identifier type="citekey">gabbasov-etal-2026-comparative</identifier>
    <location>
        <url>https://aclanthology.org/2026.docinsights-1.27/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>400</start>
            <end>408</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Comparative Study of AI Agent Architecture Efficiency in Table Processing Tasks
%A Gabbasov, Timur
%A Sandalov, Aleksandr
%A Maksimov, Aleksandr
%A Khodorchenko, Maria
%A Butakov, Nikolay Alekseevich
%Y T.Y.S.S., Santosh
%Y Agarwal, Amit
%Y Gupta, Vivek
%Y Kementchedjhieva, Yova
%Y Garimella, Aparna
%Y Hou, Yufang
%S The Proceedings of Workshop on Document Intelligence and Understanding
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-477-4
%F gabbasov-etal-2026-comparative
%X Tables are central to business document workflows, but practical table-processing tasks vary substantially in structure, content, and required action. We compare eight ReAct-agent configurations built on the same Qwen3.6-35B-A3B model. The design crosses four tool profiles - Minimal, Search, Spreadsheet, and Full - with the presence or absence of Markdown-based procedural skills. We evaluate the configurations on 40 tasks from company workflows, FinQA, and SpreadsheetBench. Across three complete replicates, comprising 960 agent-task episodes, Full with skills obtains the highest aggregate success rate of 75.0% and a mean score of 0.8061. Full without skills reaches 65.0%, while Spreadsheet without skills reaches 73.3%. Aggregated across tool profiles, skills increase success from 69.4% to 72.3%, but their effect is heterogeneous: negative for Minimal and Spreadsheet and positive for Search and Full. Overall success remains stable across replicates at 70.3%, 70.9%, and 71.3%, although individual configuration results vary more substantially. These findings suggest that procedural guidance is most promising when an agent must coordinate a broad tool environment rather than as a universal augmentation.
%U https://aclanthology.org/2026.docinsights-1.27/
%P 400-408

Markdown (Informal)

[Comparative Study of AI Agent Architecture Efficiency in Table Processing Tasks](https://aclanthology.org/2026.docinsights-1.27/) (Gabbasov et al., DocInsights 2026)

• Comparative Study of AI Agent Architecture Efficiency in Table Processing Tasks (Gabbasov et al., DocInsights 2026)

ACL

• Timur Gabbasov, Aleksandr Sandalov, Aleksandr Maksimov, Maria Khodorchenko, and Nikolay Alekseevich Butakov. 2026. Comparative Study of AI Agent Architecture Efficiency in Table Processing Tasks. In The Proceedings of Workshop on Document Intelligence and Understanding, pages 400–408, Budapest, Hungary. Association for Computational Linguistics.