Spreadsheet Automation

Follow this topic Ask about this topic

Spreadsheet automation studies how AI can interpret workbook structure and carry out tasks such as filling templates, debugging formulas, building financial models, and analyzing data. Approaches include language-model agents that plan and call spreadsheet tools, graph-based representations of relationships across cells and sheets, structured table detection, and reinforcement learning in interactive spreadsheet environments. Evaluation is moving from isolated edits toward end-to-end business and finance workflows, with measures of correctness, formatting, preservation, and professional usability. A central challenge is making agents reliable on complex tasks while keeping their actions inspectable and letting users verify or guide consequential edits.

Latest papers

20 papers4 papers in the last four weeks, with none the four weeks before. 0.0% of all new papers.

MBABench-2.0: Evaluating AI Agents on Constructing Financial Spreadsheets Beyond Simple Verifiability
Oct 22, 2026Julian Poeltl, Thomson Yen, Patrick Shea +5Spreadsheet Automation AI Agent Benchmarks
ESB Reutlingen · Columbia Business School · TensorBlock
fxBench: Evaluating and Understanding Formula Suggestions in Spreadsheets
Oct 17, 2026Sanket Mhatre, Sumit Gulwani, Vu Le +2Spreadsheet Automation Benchmark Design
Proximal · Microsoft, USA · Microsoft, India +1
CRAFT: An Agentic Spreadsheet Form Filling System with Template Awareness
Oct 3, 2026Leyao Gu, Yingjie Xiong, Zirui Tang +5Spreadsheet Automation LLM Self-Correction
Shanghai Jiao Tong University · Microsoft Research · Purdue University
The Last Mile Is the File: OfficeEditBench for Preservation-Aware Office Editing
Sep 28, 2026Zhiwen Wu, Chengxu WuBenchmark Validity Spreadsheet Automation
Peking University
Structured Prediction for Scalable Spreadsheet Table Understanding: From Cell Types to Table Ranges (Extended Version)
Aug 17, 2026Antoine Gauquier, Ioana Manolescu, Pierre SenellartTable Structure Recognition Spreadsheet Automation
DI ENS, ENS, CNRS, PSL, Inria Paris, France
SheetCompass: Hierarchical Relation Graphs for Agentic Spreadsheet Reasoning
Aug 14, 2026Panjing He, Mingyue Cheng, Yucong Luo +2Spreadsheet Automation Agentic Reasoning
State Key Laboratory of Cognitive Intelligence, University of Science and Technology of China Hefei, China
Back to the Future: A workbook time machine for spread sheet creation benchmarks
Aug 8, 2026Mansi Uniyal, Agamdeep Singh, Ananya Singha +5LLM Agent Evaluation Synthetic Benchmark Generation
Microsoft
Plans Work in Mysterious Ways: Evaluating a Plan Mode for Spreadsheet Agents
Jul 26, 2026Aayush Kumar, Avik Dutta, Sumit Gulwani +3Spreadsheet Automation Human-AI Collaboration
Microsoft *Bengaluru, India · Microsoft †Redmond, WA, USA · Microsoft ‡Cambridge, UK +1
SpreadsheetBench 2: Evaluating Agents on End-to-End Business Spreadsheet Workflows
Jun 29, 2026Jian Zhu, Yuzheng Zhang, Zeyao Ma +11Benchmark Design Spreadsheet Automation
School of Information, Renmin University of China · Aptura AI · Shortcut AI +1
A Benchmark and Framework for Evaluating Next Action Predictions in Spreadsheets
Jun 11, 2026Tejas Agrawal, Vu Le, Sumit Gulwani +1Next-Action Prediction Automated Evaluation
1Microsoft, Bangalore, India · 2Microsoft, Redmond, USA · 3Microsoft, Keerbergen, Belgium.
TabClaw: An Interactive and Self-Evolving Agent for Spreadsheet Manipulation and Table Reasoning
Jun 9, 2026Mingyue Cheng, Shuo Yu, Daoyu Wang +5Multi-Agent LLM Systems Table Reasoning
State Key Laboratory of Cognitive Intelligence, University of Science and Technology of China, Hefei, China
BlueFin: Benchmarking LLM Agents on Financial Spreadsheets
May 29, 2026Srivatsa Kundurthy, Clara Na, Colton Moraine +6LLM Agent Evaluation AI Agent Benchmarks
1Longitude Labs Inc. · 2Cornell University · 3Carnegie Mellon University
MBABench: Evaluating LLM Agents on End-to-End Spreadsheet Tasks in Finance
May 21, 2026Thomson Yen, Julian Poeltl, Harshith Srinivas Gear +9LLM Agent Evaluation AI Agent Benchmarks
Decision, Risk, and Operations Division, Columbia Business School · ESB Business School, Reutlingen University · Decision, Risk,and Operations Division, Columbia Business School
Spreadsheet-RL: Advancing Large Language Model Agents on Realistic Spreadsheet Tasks via Reinforcement Learning
May 21, 2026Banghao Chi, Yining Xie, Mingyuan Wu +9Tool-Using Agents LLM Agent Training
University of Illinois Urbana-Champaign · Meta
Spreadsheet Modeling Experiments Using GPTs on Small Problem Statements and the Wall Task
Apr 28, 2026Thomas A. Grossman, Yuan Chen, Sopiko DatuashviliML Reproducibility Spreadsheet Automation
University of San Francisco 2130 Fulton St. San Francisco, CA 94117 USA
Auditing and Controlling AI Agent Actions in Spreadsheets
Apr 22, 2026Sadra Sabouri, Zeinabsadat Saghi, Run Huang +4Human-in-the-Loop AI AI Agent Monitoring
University of Southern California Los Angeles, USA · California State Polytechnic University Pomona, USA · Microsoft Redmond, USA
Beyond Rows to Reasoning: Agentic Retrieval for Multimodal Spreadsheet Understanding and Editing
Mar 6, 2026Anmol Gulati, Sahil Sen, Waqar Sarguroh +1Agentic RAG Spreadsheet Automation
Commercial Technology and Innovation Office, PricewaterhouseCoopers, U.S.
Deterministic Preprocessing and Interpretable Fuzzy Banding for Cost-per-Student Reporting from Extracted Records
Mar 5, 2026Shane Lee, Stella NgSpreadsheet Automation
School of Computer Science, University of Technology Sydney · Management Discipline Group, University of Technology Sydney
SpreadsheetArena: Decomposing Preference in LLM Generation of Spreadsheet Workbooks
Feb 16, 2026Srivatsa Kundurthy, Clara Na, Michael Handley +5LLM Evaluation Language Model Generation Evaluation
Longitude Labs Inc., New York, NY · Cornell University, Ithaca, NY · Carnegie Mellon University, Pittsburgh, PA +1
SheetMind: Actions Set Accuracy, Agents Set the Failure Mode
Jun 14, 2025Lyuhao Chen, Xi Cheng, Yanming Kang +11Multi-Agent LLM Systems LLM Agent Evaluation
Carnegie Mellon University · Cornell University · University of California, Berkeley +2

Oct 22, 2026J/K move · Enter open · S save

MBABench-2.0: Evaluating AI Agents on Constructing Financial Spreadsheets Beyond Simple Verifiability

Julian Poeltl, Thomson Yen, Patrick Shea, Shubham Anil Varma+4

ESB Reutlingen · Columbia Business School · TensorBlock

Spreadsheet Automation AI Agent Benchmarks

Open paper AskPDF arXiv

Spreadsheets are the lingua franca of modern finance, driving valuation, forecast, audit, and many other core workflows of the industry. As LLM agents advance to automate knowledge work, spreadsheet construction is a natural and pertinent testbed. However, most existing spreadsheet benchmarks evaluate agents on fill-in-the-blank style tasks, where an agent performs minor edits in otherwise complete spreadsheets. This protocol fails to capture the most central and time-consuming part of the spreadsheet workflow: the modeling decisions, format design, and sheet layout regular knowledge workers need to toil through to build a workbook from scratch. We introduce MBABench-2.0, a benchmark of 101 original spreadsheet construction tasks, the hardest of which is estimated to take 33 hours for a modeler with 3+ years of experience. Unlike coding or math, where a unit test or a final answer can verify an agent's output, the quality of a financial spreadsheet that concerns readability, maintainability, or auditability, is not so simply verifiable. To holistically evaluate spreadsheet construction, we introduce 129 criteria spanning 12 categories that operationalize quality beyond the accuracy of the spreadsheet. Evaluating over 20 agents across 8 models and 4 harness types, we find that frontier agents rarely construct spreadsheets that are both accurate and meet our quality criteria: the top agent achieves a pass rate of only 16% on the benchmark. The benchmark demonstrates a substantial gap in spreadsheet building capability of current agents.