SpreadsheetBench: Towards Challenging Real World Spreadsheet Manipulation

TL;DR

SpreadsheetBench is a new benchmark derived from real-world spreadsheet questions, designed to evaluate large language models' ability to handle complex, practical spreadsheet tasks with diverse data and multiple test cases.

Contribution

It introduces a challenging, real-world-based benchmark with a novel evaluation metric, highlighting the gap between current LLMs and human performance in spreadsheet manipulation.

Findings

Current LLMs perform significantly worse than humans on the benchmark.

The benchmark includes diverse real-world spreadsheet tasks from online forums.

Evaluation method ensures robustness across varying spreadsheet data.

Abstract

We introduce SpreadsheetBench, a challenging spreadsheet manipulation benchmark exclusively derived from real-world scenarios, designed to immerse current large language models (LLMs) in the actual workflow of spreadsheet users. Unlike existing benchmarks that rely on synthesized queries and simplified spreadsheet files, SpreadsheetBench is built from 912 real questions gathered from online Excel forums, which reflect the intricate needs of users. The associated spreadsheets from the forums contain a variety of tabular data such as multiple tables, non-standard relational tables, and abundant non-textual elements. Furthermore, we propose a more reliable evaluation metric akin to online judge platforms, where multiple spreadsheet files are created as test cases for each instruction, ensuring the evaluation of robust solutions capable of handling spreadsheets with varying values. Our comprehensive evaluation of various LLMs under both single-round and multi-round inference settings reveals a substantial gap between the state-of-the-art (SOTA) models and human performance, highlighting the benchmark's difficulty.