SheetCopilot: Bringing Software Productivity to the Next Level through Large Language Models
This addresses the problem of repetitive and error-prone spreadsheet work for end users who lack automation skills, representing an incremental improvement in software productivity.
The authors tackled automating spreadsheet tasks with natural language by proposing SheetCopilot, an agent that uses large language models to control spreadsheets, achieving a 44.3% task completion rate in a single generation.
Computer end users have spent billions of hours completing daily tasks like tabular data processing and project timeline scheduling. Most of these tasks are repetitive and error-prone, yet most end users lack the skill to automate these burdensome works. With the advent of large language models (LLMs), directing software with natural language user requests become a reachable goal. In this work, we propose a SheetCopilot agent that takes natural language task and control spreadsheet to fulfill the requirements. We propose a set of atomic actions as an abstraction of spreadsheet software functionalities. We further design a state machine-based task planning framework for LLMs to robustly interact with spreadsheets. We curate a representative dataset containing 221 spreadsheet control tasks and establish a fully automated evaluation pipeline for rigorously benchmarking the ability of LLMs in software control tasks. Our SheetCopilot correctly completes 44.3\% of tasks for a single generation, outperforming the strong code generation baseline by a wide margin. Our project page:https://sheetcopilot.github.io/.