@inproceedings{zhu-etal-2026-benchmarking,
title = "Benchmarking Aggregation Queries over Unstructured Text",
author = "Zhu, Haojia and
Jin, Jiahui and
Li, Haoyu and
Xu, Qinyuan and
Liu, Yuxi and
Zhou, Zihan and
Qiu, Hanchen and
Chen, Jiaoyan and
Zhang, Jinghui",
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.1026/",
pages = "23591--23615",
ISBN = "979-8-89176-459-0",
abstract = "Aggregation query over free text is a long-standing yet underexplored problem. It requires exhaustive evidence collection, where systems must ``find all'' satisfying entities rather than merely ``find one'' plausible answer. Existing paradigms are not designed for this completeness requirement, leading to incomplete retrieval and systematic undercounting. This limitation stems from the lack of suitable benchmarks. In this work, we formalize corpus-bounded entity-level aggregation querying over text, where answers must be grounded in the given text corpus. We introduce \textsc{AGGBench}, an aggregation benchmark constructed from finance reports and research papers through a corpus-grounded pipeline with LLM-assisted human validation. \textsc{AGGBench} contains 498 aggregation queries, including both single- and multi-constraint queries with evidence-supported gold answers. Experiments on representative basic RAG, structured and schema-first and agentic methods show that existing systems struggle to preserve evidence coverage and accurately count satisfying entities, highlighting the need for completeness-oriented methods for aggregation over unstructured text. The data and code are available in \url{https://github.com/luck-seu/AGGBench}."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhu-etal-2026-benchmarking">
<titleInfo>
<title>Benchmarking Aggregation Queries over Unstructured Text</title>
</titleInfo>
<name type="personal">
<namePart type="given">Haojia</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiahui</namePart>
<namePart type="family">Jin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haoyu</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qinyuan</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuxi</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zihan</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hanchen</namePart>
<namePart type="family">Qiu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiaoyan</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jinghui</namePart>
<namePart type="family">Zhang</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>Aggregation query over free text is a long-standing yet underexplored problem. It requires exhaustive evidence collection, where systems must “find all” satisfying entities rather than merely “find one” plausible answer. Existing paradigms are not designed for this completeness requirement, leading to incomplete retrieval and systematic undercounting. This limitation stems from the lack of suitable benchmarks. In this work, we formalize corpus-bounded entity-level aggregation querying over text, where answers must be grounded in the given text corpus. We introduce AGGBench, an aggregation benchmark constructed from finance reports and research papers through a corpus-grounded pipeline with LLM-assisted human validation. AGGBench contains 498 aggregation queries, including both single- and multi-constraint queries with evidence-supported gold answers. Experiments on representative basic RAG, structured and schema-first and agentic methods show that existing systems struggle to preserve evidence coverage and accurately count satisfying entities, highlighting the need for completeness-oriented methods for aggregation over unstructured text. The data and code are available in https://github.com/luck-seu/AGGBench.</abstract>
<identifier type="citekey">zhu-etal-2026-benchmarking</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.1026/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>23591</start>
<end>23615</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Benchmarking Aggregation Queries over Unstructured Text
%A Zhu, Haojia
%A Jin, Jiahui
%A Li, Haoyu
%A Xu, Qinyuan
%A Liu, Yuxi
%A Zhou, Zihan
%A Qiu, Hanchen
%A Chen, Jiaoyan
%A Zhang, Jinghui
%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 zhu-etal-2026-benchmarking
%X Aggregation query over free text is a long-standing yet underexplored problem. It requires exhaustive evidence collection, where systems must “find all” satisfying entities rather than merely “find one” plausible answer. Existing paradigms are not designed for this completeness requirement, leading to incomplete retrieval and systematic undercounting. This limitation stems from the lack of suitable benchmarks. In this work, we formalize corpus-bounded entity-level aggregation querying over text, where answers must be grounded in the given text corpus. We introduce AGGBench, an aggregation benchmark constructed from finance reports and research papers through a corpus-grounded pipeline with LLM-assisted human validation. AGGBench contains 498 aggregation queries, including both single- and multi-constraint queries with evidence-supported gold answers. Experiments on representative basic RAG, structured and schema-first and agentic methods show that existing systems struggle to preserve evidence coverage and accurately count satisfying entities, highlighting the need for completeness-oriented methods for aggregation over unstructured text. The data and code are available in https://github.com/luck-seu/AGGBench.
%U https://aclanthology.org/2026.emnlp-main.1026/
%P 23591-23615
Markdown (Informal)
[Benchmarking Aggregation Queries over Unstructured Text](https://aclanthology.org/2026.emnlp-main.1026/) (Zhu et al., EMNLP 2026)
ACL
- Haojia Zhu, Jiahui Jin, Haoyu Li, Qinyuan Xu, Yuxi Liu, Zihan Zhou, Hanchen Qiu, Jiaoyan Chen, and Jinghui Zhang. 2026. Benchmarking Aggregation Queries over Unstructured Text. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 23591–23615, Budapest, Hungary. Association for Computational Linguistics.