@inproceedings{dong-etal-2024-encoding,
title = "Encoding Spreadsheets for Large Language Models",
author = "Dong, Haoyu and
Zhao, Jianbo and
Tian, Yuzhang and
Xiong, Junyu and
Zhou, Mengyu and
Lin, Yun and
Cambronero, Jos{\'e} and
He, Yeye and
Han, Shi and
Zhang, Dongmei",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.1154",
pages = "20728--20748",
abstract = "Spreadsheets are characterized by their extensive two-dimensional grids, flexible layouts, and varied formatting options, which pose significant challenges for large language models (LLMs). In response, we introduce SheetEncoder, pioneering an efficient encoding method designed to unleash and optimize LLMs{'} powerful understanding and reasoning capability on spreadsheets. Initially, we propose a vanilla serialization approach that incorporates cell addresses, values, and formats. However, this approach was limited by LLMs{'} token constraints, making it impractical for most applications. To tackle this challenge, three innovative modules are proposed to compress spreadsheets effectively: structural-anchor-based compression, inverse index translation, and data-format-aware aggregation. It significantly improves performance in spreadsheet table detection task, outperforming the vanilla approach by 25.6{\%} in GPT4{'}s in-context learning setting. Moreover, fine-tuned LLM with SheetEncoder has an average compression ratio of 25{\mbox{$\times$}}, but achieves a state-of-the-art 78.9{\%} F1 score, surpassing the best existing models by 12.3{\%}, demonstrating that SheetEncoder greatly boosts LLMs{'}s performance on spreadsheet data.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="dong-etal-2024-encoding">
<titleInfo>
<title>Encoding Spreadsheets for Large Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Haoyu</namePart>
<namePart type="family">Dong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jianbo</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuzhang</namePart>
<namePart type="family">Tian</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Junyu</namePart>
<namePart type="family">Xiong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mengyu</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yun</namePart>
<namePart type="family">Lin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">José</namePart>
<namePart type="family">Cambronero</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yeye</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shi</namePart>
<namePart type="family">Han</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dongmei</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yaser</namePart>
<namePart type="family">Al-Onaizan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohit</namePart>
<namePart type="family">Bansal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yun-Nung</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Miami, Florida, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Spreadsheets are characterized by their extensive two-dimensional grids, flexible layouts, and varied formatting options, which pose significant challenges for large language models (LLMs). In response, we introduce SheetEncoder, pioneering an efficient encoding method designed to unleash and optimize LLMs’ powerful understanding and reasoning capability on spreadsheets. Initially, we propose a vanilla serialization approach that incorporates cell addresses, values, and formats. However, this approach was limited by LLMs’ token constraints, making it impractical for most applications. To tackle this challenge, three innovative modules are proposed to compress spreadsheets effectively: structural-anchor-based compression, inverse index translation, and data-format-aware aggregation. It significantly improves performance in spreadsheet table detection task, outperforming the vanilla approach by 25.6% in GPT4’s in-context learning setting. Moreover, fine-tuned LLM with SheetEncoder has an average compression ratio of 25\times, but achieves a state-of-the-art 78.9% F1 score, surpassing the best existing models by 12.3%, demonstrating that SheetEncoder greatly boosts LLMs’s performance on spreadsheet data.</abstract>
<identifier type="citekey">dong-etal-2024-encoding</identifier>
<location>
<url>https://aclanthology.org/2024.emnlp-main.1154</url>
</location>
<part>
<date>2024-11</date>
<extent unit="page">
<start>20728</start>
<end>20748</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Encoding Spreadsheets for Large Language Models
%A Dong, Haoyu
%A Zhao, Jianbo
%A Tian, Yuzhang
%A Xiong, Junyu
%A Zhou, Mengyu
%A Lin, Yun
%A Cambronero, José
%A He, Yeye
%A Han, Shi
%A Zhang, Dongmei
%Y Al-Onaizan, Yaser
%Y Bansal, Mohit
%Y Chen, Yun-Nung
%S Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing
%D 2024
%8 November
%I Association for Computational Linguistics
%C Miami, Florida, USA
%F dong-etal-2024-encoding
%X Spreadsheets are characterized by their extensive two-dimensional grids, flexible layouts, and varied formatting options, which pose significant challenges for large language models (LLMs). In response, we introduce SheetEncoder, pioneering an efficient encoding method designed to unleash and optimize LLMs’ powerful understanding and reasoning capability on spreadsheets. Initially, we propose a vanilla serialization approach that incorporates cell addresses, values, and formats. However, this approach was limited by LLMs’ token constraints, making it impractical for most applications. To tackle this challenge, three innovative modules are proposed to compress spreadsheets effectively: structural-anchor-based compression, inverse index translation, and data-format-aware aggregation. It significantly improves performance in spreadsheet table detection task, outperforming the vanilla approach by 25.6% in GPT4’s in-context learning setting. Moreover, fine-tuned LLM with SheetEncoder has an average compression ratio of 25\times, but achieves a state-of-the-art 78.9% F1 score, surpassing the best existing models by 12.3%, demonstrating that SheetEncoder greatly boosts LLMs’s performance on spreadsheet data.
%U https://aclanthology.org/2024.emnlp-main.1154
%P 20728-20748
Markdown (Informal)
[Encoding Spreadsheets for Large Language Models](https://aclanthology.org/2024.emnlp-main.1154) (Dong et al., EMNLP 2024)
ACL
- Haoyu Dong, Jianbo Zhao, Yuzhang Tian, Junyu Xiong, Mengyu Zhou, Yun Lin, José Cambronero, Yeye He, Shi Han, and Dongmei Zhang. 2024. Encoding Spreadsheets for Large Language Models. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 20728–20748, Miami, Florida, USA. Association for Computational Linguistics.