中文

SpreadsheetBench 2: Evaluating Agents on End-to-End Business Spreadsheet Workflows

软件工程 2026-06-29 v1 人工智能

摘要

Spreadsheets are widely used for business analysis, financial modeling, reporting, and decision-making. However, most existing spreadsheet benchmarks evaluate isolated operations such as single-formula generation or local cell edits, and therefore fail to capture end-to-end workflows in realistic business settings. We introduce \textsc{SpreadsheetBench 2}, a workflow-level benchmark for spreadsheet agents that covers three task categories: generation, debugging, and visualization. The benchmark is constructed from authentic business data, including financial reports and corporate filings, and is annotated and validated by domain experts. The benchmark contains 321 tasks; each instance averages 11.8 worksheets and requires 593.5 cell modifications, reflecting large multi-sheet workbooks with cross-sheet dependencies. We evaluate eight frontier large language models under a unified multi-turn agent scaffold, and additionally include several LLM-based spreadsheet products as complementary baselines. Results show that current systems remain far from reliable on real-world workflows: the best model achieves 34.89\% overall task accuracy, and debugging accuracy is as low as 12.00\%. Trajectory analysis and a failure taxonomy further indicate that insufficient spreadsheet inspection and incorrect target-cell selection are the dominant bottlenecks. Together, these findings position \textsc{SpreadsheetBench 2} as a challenging testbed for advancing reliable spreadsheet automation. Project page: https://spreadsheetbench.github.io/

引用

@article{arxiv.2606.29955,
  title  = {SpreadsheetBench 2: Evaluating Agents on End-to-End Business Spreadsheet Workflows},
  author = {Jian Zhu and Yuzheng Zhang and Zeyao Ma and Bohan Zhang and Armin Schoepf and Daniel Woloch and Peter Yiliu Wang and Guangyu Robert Yang and Samuel Jacob and Siddharth Nagisetty and Abhiram Chundru and Jean Lin and Spencer Mateega and Jing Zhang},
  journal= {arXiv preprint arXiv:2606.29955},
  year   = {2026}
}