SEAIJun 29

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

arXiv:2606.2995516.9
Predicted impact top 15% in SE · last 90 daysOriginality Synthesis-oriented
AI Analysis

For researchers and developers of spreadsheet automation agents, this benchmark reveals that existing LLMs struggle with real-world multi-sheet workflows, highlighting specific bottlenecks like insufficient inspection and incorrect cell selection.

SpreadsheetBench 2 introduces a workflow-level benchmark for spreadsheet agents, covering generation, debugging, and visualization tasks from authentic business data. The best model achieves 34.89% overall accuracy, with debugging accuracy as low as 12.00%, showing current systems are far from reliable.

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/

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes