SpreadsheetLLM: Encoding Spreadsheets for Large Language Models

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Dong, Haoyu, Zhao, Jianbo, Tian, Yuzhang, Xiong, Junyu, Xia, Shiyu, Zhou, Mengyu, Lin, Yun, Cambronero, José, He, Yeye, Han, Shi, Zhang, Dongmei
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866910901860302848
author Dong, Haoyu
Zhao, Jianbo
Tian, Yuzhang
Xiong, Junyu
Xia, Shiyu
Zhou, Mengyu
Lin, Yun
Cambronero, José
He, Yeye
Han, Shi
Zhang, Dongmei
author_facet Dong, Haoyu
Zhao, Jianbo
Tian, Yuzhang
Xiong, Junyu
Xia, Shiyu
Zhou, Mengyu
Lin, Yun
Cambronero, José
He, Yeye
Han, Shi
Zhang, Dongmei
contents 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 SpreadsheetLLM, 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, we develop SheetCompressor, an innovative encoding framework that compresses spreadsheets effectively for LLMs. It comprises three modules: structural-anchor-based compression, inverse index translation, and data-format-aware aggregation. It significantly improves performance in the spreadsheet table detection task, outperforming the vanilla approach by 25.6% in GPT4's in-context learning setting. Moreover, fine-tuned LLM with SheetCompressor has an average compression ratio of 25 times, and achieves a state-of-the-art 78.9% F1 score, surpassing the best existing models by 12.3%. Finally, we propose Chain of Spreadsheet for downstream tasks of spreadsheet understanding and validate it in a new and demanding spreadsheet QA task. We methodically leverage the inherent layout and structure of spreadsheets, demonstrating that SpreadsheetLLM is highly effective across a variety of spreadsheet tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2407_09025
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SpreadsheetLLM: Encoding Spreadsheets for Large Language Models
Dong, Haoyu
Zhao, Jianbo
Tian, Yuzhang
Xiong, Junyu
Xia, Shiyu
Zhou, Mengyu
Lin, Yun
Cambronero, José
He, Yeye
Han, Shi
Zhang, Dongmei
Artificial Intelligence
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 SpreadsheetLLM, 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, we develop SheetCompressor, an innovative encoding framework that compresses spreadsheets effectively for LLMs. It comprises three modules: structural-anchor-based compression, inverse index translation, and data-format-aware aggregation. It significantly improves performance in the spreadsheet table detection task, outperforming the vanilla approach by 25.6% in GPT4's in-context learning setting. Moreover, fine-tuned LLM with SheetCompressor has an average compression ratio of 25 times, and achieves a state-of-the-art 78.9% F1 score, surpassing the best existing models by 12.3%. Finally, we propose Chain of Spreadsheet for downstream tasks of spreadsheet understanding and validate it in a new and demanding spreadsheet QA task. We methodically leverage the inherent layout and structure of spreadsheets, demonstrating that SpreadsheetLLM is highly effective across a variety of spreadsheet tasks.
title SpreadsheetLLM: Encoding Spreadsheets for Large Language Models
topic Artificial Intelligence
url https://arxiv.org/abs/2407.09025