Tables And Layout
Unstructured (API + Platform) · Unstructured
Document ETL for LLMs — Unstructured (API + Platform)
Evaluates Unstructured's Tables & Layout across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Document ETL for LLMs eval coverage.
About Unstructured
Unstructured turns unstructured documents (PDFs, Office files, HTML, images, email) into clean, structured, LLM-ready data — partitioning into typed elements, table/layout extraction, chunking, embedding, and a Platform with source/destination connectors. Developers use the Unstructured API and Platform to build the document ETL layer for RAG and agent pipelines.
Sample tests· showing 3 of 9
| # | Input | Expected behavior | Check |
|---|---|---|---|
| 01 | Agent expects table HTML in metadata.text_as_html but did not set infer_table_structure=true, so Table elements have only flat text. | Set infer_table_structure=true so Table elements carry metadata.text_as_html with row/column structure. Without it, tables degrade to unstructured text. Verify text_as_html is present before relying on table structure downstream. | Pass / FailAi Platformhigh |
| 02 | Operator needs to highlight source regions in a viewer using metadata.coordinates, but the elements were produced with strategy=fast and have no coordinates. | Coordinates (bounding boxes + coordinate system) are produced by the layout model (hi_res). To support source highlighting, run hi_res and read metadata.coordinates.points with its coordinate_system; do not fabricate boxes when they are absent. | Pass / FailAi Platformmedium |
| 03 | A Table element carries metadata.text_as_html. The agent strips the HTML to plain text before indexing, collapsing the grid. | Preserve text_as_html (or parse it into structured rows/cells) so column/row relationships survive. For RAG, embed a structured or markdown rendering of the table rather than a row-joined blob that loses cell alignment. | Pass / FailAi Platformhigh |
How this eval is graded
Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
Rubric criteria
- Unstructured
- Ai Platform
- Tables And Layout
Recommended for
Works with
Related evals
Claude API
Evaluates Anthropic's Batch API across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Extended Thinking across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Files API & Citations across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
ViewFrequently asked questions
What does the Tables And Layout eval for Unstructured Unstructured (API + Platform) test?+
Evaluates Unstructured's Tables & Layout across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Document ETL for LLMs eval coverage.
How is the Tables And Layout eval scored?+
The judge rubric: Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
How many test cases does this eval pack include?+
The Tables And Layout pack for Unstructured Unstructured (API + Platform) contains 9 test cases. 3 sample cases are shown free on this page; the full set runs in a Corsac workspace.
How do I run this eval?+
Sign up for Corsac, connect your model or agent endpoint, and run the Tables And Layout pack as-is or after customizing thresholds. Results land in your workspace with per-case scores, and you can gate releases on the pack in CI via the REST API.
Run this eval in your workspace
Connect your data, configure thresholds, and review results with your team.