OfficeEditBench measures accuracy in maintaining complex office file edits
The Last Mile Is the File: OfficeEditBench for Preservation-Aware Office Editing
Software EngineeringComputation and Language
Summary
Editing office files like spreadsheets or documents can be tricky because small changes may need careful updates without breaking other parts. The authors present OfficeEditBench, a set of 170 editing tasks that test how well tools can make required updates while keeping the rest unchanged. Their study shows that even when tools deliver valid files, they often fail to fully meet all editing requirements, such as preserving formulas or interrelated rules. This highlights why simply fixing one part of a file isn't enough to ensure the whole file stays consistent and usable.
What this means in practice
- •For office software developers: Evaluate and improve the accuracy of automatic edits to spreadsheets and documents to preserve intended content and dependencies.
- •For enterprise it support teams: Test office file maintenance tools to ensure changes keep organizational documents consistent and avoid unintended data corruption.
Authors
Zhiwen Wu, Chengxu Wu
Abstract
A small Office edit creates two obligations: propagate every required update and leave protected state untouched. Updating too little leaves dependencies inconsistent; updating too much changes content the user did not authorize. We introduce OfficeEditBench, a 170-task benchmark for change-scoped maintenance of spreadsheets, presentations, and documents. Task contracts specify required updates, protected state, native structures, and applicable interaction requirements. Across 510 archived task-system outcomes from WorkBuddy, Doubao, and Codex, we distinguish file delivery, target completion, and verifier-defined acceptance. Hard package-valid delivery ranges from 92% to 100%, yet no selected output satisfies the complete contract. Case analysis highlights why local correctness is insufficient: an updated value can lose its generating formula, a revised rule can fail to reach related conclusions, and a new deadline can omit a retained prerequisite. These mechanisms connect artifact-level checks to the continued maintainability of Office files. We analyze maintenance failures while distinguishing frozen automatic verdicts from human acceptability. OfficeEditBench provides a testbed for completing required changes while preserving the logic and scope of existing work.