@inproceedings{wu-etal-2026-boosting,
title = "Boosting Text-to-Chart Retrieval through Synthesized Hierarchical Insights",
author = "Wu, Yifan and
Yan, Lutao and
Zhu, Yizhang and
Tseng, Yenchi and
Wang, Yong and
Wang, Jiannan and
Tang, Nan and
Luo, Yuyu",
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.2098/",
pages = "47764--47780",
ISBN = "979-8-89176-459-0",
abstract = "Text-to-chart retrieval enables users to locate relevant charts through natural language queries and has emerged as a critical capability in modern business intelligence systems. However, current benchmarks fail to simulate realistic user queries or test deep semantic understanding, leaving a significant evaluation gap. To address this, we introduce CRBench, the first real-world BI-sourced benchmark comprising 21,862 charts and 326 queries, which adopts a Target-and-Distractor paradigm to evaluate discriminative retrieval among highly similar candidates. Benchmarking reveals that existing methods, which rely primarily on visual features, fail to capture the rich analytical semantics of charts. We therefore propose a semantic insights synthesis pipeline that automatically generates three hierarchical levels of insights for alignment training: visual patterns, statistical properties, and practical applications. Using this pipeline, we produce 207,498 insights for 69,166 charts as training data. Leveraging this data, we develop ChartFinder, which significantly outperforms state-of-the-art methods on CRBench, achieving 66.9{\%} NDCG@10 for precise queries (an 11.58{\%} improvement) and an average gain of 5{\%} across fuzzy query metrics."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wu-etal-2026-boosting">
<titleInfo>
<title>Boosting Text-to-Chart Retrieval through Synthesized Hierarchical Insights</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yifan</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lutao</namePart>
<namePart type="family">Yan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yizhang</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yenchi</namePart>
<namePart type="family">Tseng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yong</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiannan</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nan</namePart>
<namePart type="family">Tang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuyu</namePart>
<namePart type="family">Luo</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>Text-to-chart retrieval enables users to locate relevant charts through natural language queries and has emerged as a critical capability in modern business intelligence systems. However, current benchmarks fail to simulate realistic user queries or test deep semantic understanding, leaving a significant evaluation gap. To address this, we introduce CRBench, the first real-world BI-sourced benchmark comprising 21,862 charts and 326 queries, which adopts a Target-and-Distractor paradigm to evaluate discriminative retrieval among highly similar candidates. Benchmarking reveals that existing methods, which rely primarily on visual features, fail to capture the rich analytical semantics of charts. We therefore propose a semantic insights synthesis pipeline that automatically generates three hierarchical levels of insights for alignment training: visual patterns, statistical properties, and practical applications. Using this pipeline, we produce 207,498 insights for 69,166 charts as training data. Leveraging this data, we develop ChartFinder, which significantly outperforms state-of-the-art methods on CRBench, achieving 66.9% NDCG@10 for precise queries (an 11.58% improvement) and an average gain of 5% across fuzzy query metrics.</abstract>
<identifier type="citekey">wu-etal-2026-boosting</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.2098/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>47764</start>
<end>47780</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Boosting Text-to-Chart Retrieval through Synthesized Hierarchical Insights
%A Wu, Yifan
%A Yan, Lutao
%A Zhu, Yizhang
%A Tseng, Yenchi
%A Wang, Yong
%A Wang, Jiannan
%A Tang, Nan
%A Luo, Yuyu
%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 wu-etal-2026-boosting
%X Text-to-chart retrieval enables users to locate relevant charts through natural language queries and has emerged as a critical capability in modern business intelligence systems. However, current benchmarks fail to simulate realistic user queries or test deep semantic understanding, leaving a significant evaluation gap. To address this, we introduce CRBench, the first real-world BI-sourced benchmark comprising 21,862 charts and 326 queries, which adopts a Target-and-Distractor paradigm to evaluate discriminative retrieval among highly similar candidates. Benchmarking reveals that existing methods, which rely primarily on visual features, fail to capture the rich analytical semantics of charts. We therefore propose a semantic insights synthesis pipeline that automatically generates three hierarchical levels of insights for alignment training: visual patterns, statistical properties, and practical applications. Using this pipeline, we produce 207,498 insights for 69,166 charts as training data. Leveraging this data, we develop ChartFinder, which significantly outperforms state-of-the-art methods on CRBench, achieving 66.9% NDCG@10 for precise queries (an 11.58% improvement) and an average gain of 5% across fuzzy query metrics.
%U https://aclanthology.org/2026.emnlp-main.2098/
%P 47764-47780
Markdown (Informal)
[Boosting Text-to-Chart Retrieval through Synthesized Hierarchical Insights](https://aclanthology.org/2026.emnlp-main.2098/) (Wu et al., EMNLP 2026)
ACL
- Yifan Wu, Lutao Yan, Yizhang Zhu, Yenchi Tseng, Yong Wang, Jiannan Wang, Nan Tang, and Yuyu Luo. 2026. Boosting Text-to-Chart Retrieval through Synthesized Hierarchical Insights. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 47764–47780, Budapest, Hungary. Association for Computational Linguistics.