SquarePact Research · Legal AI Benchmark

The SquarePact Formatting Benchmark

A public dataset that measures one thing precisely: can an AI edit a real contract in Word, defined terms, numbered clauses and redlines included, without breaking the formatting a legal team depends on?

The same AI edit to a contract clause shown with formatting preserved and with formatting broken
The same AI edit to a contract clause: formatting preserved versus the drift and broken structure a legal document cannot ship with. This distinction is what the benchmark measures.

Introduction

Today we are releasing the SquarePact Formatting Benchmark, a public dataset for evaluating how well AI tools edit real Word documents without damaging their formatting. To our knowledge, it is the first widely available benchmark of its kind.

The last few years brought real advances in legal and professional AI, and a wave of benchmarks to measure it. Most score reasoning, retrieval, drafting, or citation accuracy. Almost none ask the question that decides whether a tool can be trusted on a live document: can it make the change you asked for without breaking everything around it?

Ask a general-purpose assistant to revise one clause and the output often arrives with a broken numbered list, a font that drifted, a heading that lost its style, or track-changes that no longer make sense. For the documents SquarePact is built for, such as contracts, proposals, reports, and policies, this is not cosmetic. It is the difference between shipping a document and re-doing the work by hand.

We built this benchmark to measure that capability directly, and we are publishing the dataset so anyone can inspect it, reproduce our procedure, and test their own tools against it.

TL;DR / FAQ

In short: we are releasing the first public dataset for formatting-preserving AI edits in Word, comparing four Word add-ins. It keeps only the cases where tools measurably differ, is free to download with attribution, and is versioned over time. Scores live in Section 8.

Why formatting is the missing benchmark

Formatting-preserving editing is under-measured for three distinct reasons. Existing resources tend to fail on at least one of them.

Capability mismatch

Most benchmarks reward answer quality or drafting fluency. They rarely test whether an edit preserves the structure and formatting of a document that already exists, which is exactly where professional tools tend to break.

Lack of transparency

The few datasets that do touch this capability are not openly available. They cannot be downloaded, inspected, or independently verified, so claims about them cannot be checked.

No adaptivity

Static benchmarks go stale as soon as tools optimize against them, and nothing keeps pace as the underlying systems improve. A benchmark for a moving target has to move too.

The dataset described here is built to address all three: it is public, inspectable, and versioned over time.

Systems evaluated

We evaluate four add-ins that operate inside Microsoft Word, where the editing work actually happens. Each receives an identical edit instruction on an identical document portion.

  • SquarePact OursWord add-in · tested Jul 2026
  • Microsoft CopilotWord integration · tested Jul 2026
  • Claude for WordWord add-in · tested Jul 2026
  • Grok for WordWord add-in · tested Jul 2026
Table 1. Systems included in v1.0. Exact build identifiers are recorded in the released metadata.

What we test

Each edited portion is scored on five formatting-preservation measures. Every measure targets a concrete way a document degrades when an edit is applied carelessly.

Measure What it checks Example failure
Character & paragraph formattingRun- and paragraph-level properties survive the edit.Pasted text drops to Calibri 11 mid-paragraph.
Styles & headingsNamed styles and heading levels persist.A "Heading 2" silently becomes body text.
Tables & list structureCell boundaries and list nesting hold.A three-level list collapses to a single level.
Numbering & cross-referencesAutomatic numbering and references stay valid."See Section 4" now points to the wrong clause.
Track changes & commentsRevision marks and comments survive the edit.An accepted change wipes an open comment thread.
Table 2. The five measures, what each verifies, and a representative failure mode.

How we built it

The construction procedure is deliberately simple so that it can be reproduced. Some specifics are being finalized for the v1.0 write-up and are marked below.

  1. Source corpus

    We begin with a subset of the SuperDoc docx-corpus, an openly available collection of real Word documents. See github.com/superdoc-dev/docx-corpus.

  2. Random sampling Details to follow

    From that corpus we draw a random subset of documents, avoiding selection that would favor any particular system.

  3. Portion selection Details to follow

    Within each sampled document we select specific portions to edit, isolating the formatting features under test.

  4. Standard tests Details to follow

    We apply a standard set of editing tests to each portion, run identically across every system in the comparison.

  5. Scoring & reporting

    Each portion is scored on the five measures in Section 5, and results are reported in Section 8. Only portions that satisfy the criterion in Section 7 are retained.

The difference-only criterion

This benchmark does not aim to be exhaustive. Because its purpose is to improve tools, we retain a document portion only when the evaluated systems measurably differ on at least one measure. Portions on which every system behaves identically are excluded: they add size without adding signal.

portion A styles diverged across systems Retained
portion B list numbering broke in 2 of 4 Retained
portion C all systems identical Excluded
portion D track-changes lost in 1 of 4 Retained
portion E all systems identical Excluded
Figure 1. The difference-only criterion in action. What remains is a concentrated set of the cases where formatting-preservation genuinely breaks down.

Results

Table 3 reports per-measure scores for each system on the v1.0 set, with the composite as the mean across measures. Higher is better. Full figures publish with the dataset; each score is dated and tied to the build identifiers in the released metadata.

Measure SquarePact Copilot Claude (Word) Grok (Word)
Character & paragraph formatting
Styles & headings
Tables & list structure
Numbering & cross-references
Track changes & comments
Composite
Table 3. Per-measure scores on the v1.0 set. The SquarePact column is highlighted for reference only.

Access & license

The release bundles the document portions, the tests applied to each, and the scored results, so anyone can replicate our numbers or evaluate a system we did not include.

Contents
Source .docx portions, per-portion test definitions, scored results, methodology write-up.
Version
1.0 (July 2026)
License
Permissive. Free to use for research or product work; attribution and a link back to this page are required.
Corpus
Subset of the SuperDoc docx-corpus

Download v1.0Available on release.

Built a tool you would like measured against this set? The full procedure is public, so you can run it yourself, or see what the dataset powers and try SquarePact free in Word.

Versioning & maintenance

We expect systems to improve against this benchmark over time, including by optimizing directly for it. That is a sign the benchmark is doing its job. To keep it meaningful, every test is dated, results are tied to a version, and new cases are added on a regular cadence. Reported numbers always reference a specific version and date.

Citation

If you use this benchmark, please cite it and link back to this page.

@techreport{squarepact_formatting_benchmark_2026,
  title  = {The SquarePact Formatting Benchmark},
  author = {Licato, John and Nighojkar, Animesh and Vaidya, Darsh and Sajithkumar, Govind},
  year   = {2026},
  note   = {Version 1.0},
  institution = {Actualization AI, Inc.},
  url    = {https://squarepact.com/formattingBench/}
}

Get notified when v1.0 is released.

We will send the dataset link and future versions as the benchmark grows. No spam.