Does LiteParse keep mathematical equations as math markup or an honest fallback?
LiteParse turned the note's equations into scrambled text: its Markdown has no math markup and no placeholder, each display equation is split across lines in the wrong order, the ω of equation (4)'s ω/c factor is missing, and the top of equation (6) became a heading.
1 of 1 test case failed
Every test case in this scenario has a result.
The test case
Each test case is judged on its own: Pass, Fail, Not gradable, or Untested. The scenario result above is based on this test case.
Equations — math markup or honest fallbackFailEvidence
We called LiteParse 2.14.7 in Python on the complete three-page technical note with the default settings and Markdown output.
Copied from Output file (Markdown), lines 17–23
∂A ∂Q + = 0 (1) ∂t ∂x
Copied from Output file (Markdown), lines 49–51
*c ( χ* ))1/2 8 n 2 *c η0* *μ =* , *χ =* 2(1 + χ 2 3π h 4/3 (4)
Copied from Output file (Markdown), lines 63–67
*b n(h) = n∞ +* (5) *h 1/3*
Copied from Output file (Markdown), lines 77–81
## |ηo - ηm| *ε =* *ηo* (6)
Returned Markdown: the equations as scattered text fragments with no math markup
Open the Markdown file ↗| ✗ | Expected: Equations come out as math markup, such as LaTeX or MathML, or an honest fallback, not garbled prose.Found: No math markup or placeholder; every display equation is split over several out-of-order lines, equation (4) drops the ω from its ω/c factor, and equation (6)'s numerator became a heading. |
The test needed math markup or an honest fallback instead of garbled prose. The output has neither: equations are broken into out-of-order fragments, equation (4) drops the ω from its ω/c factor, and part of equation (6) is a heading.
Prose sections intact. The prose sections came through well: section headings became Markdown headings and the paragraphs read correctly.
Inline symbols displaced. Inline symbols are displaced too: in the paragraph after equation (3), the symbols are lifted out of the sentence that defines them.
Configuration and setup
How this tool was set up for the run and what the test needed in place. Each row is a fact from the run's records; a fact the records do not hold is left out, not guessed.
How this scenario is graded
How we decide Pass, Fail and Not gradable. The same rules apply to every tool tested on this scenario.
Each test case gets one result per tool: Pass, Fail or Not gradable. A test case we haven't run yet shows Untested. There are no partial results.
- Pass: the tool did everything the test expected, and nothing it said contradicts the correct answer.
- Fail: at least one expected behaviour clearly didn't happen; the test case's result says which one and quotes the tool.
- Not gradable: our evidence couldn't settle the outcome (for example a record we needed is missing). It is never counted as a fail, and the test case's result says what's missing.
Where this sits in the benchmark
This page is one cell of a larger study: one tool, one scenario. Only this benchmark's frame appears here.
| Level | Name | Scope |
|---|---|---|
| Benchmark | Converting a complex PDF into clean Markdown with an open-source library → | 12 scenarios · 10 tools |
| Capability | Equations & Mathematical Notation → | |
| Scenario | A document containing mathematical equations → | |
| Tool | LiteParse → |
Global scenario definition → · Global capability definition → · LiteParse product page →
History of this result
What has happened to this result since it was first published. Runs and grades are never overwritten: a retest or a re-grade publishes a new result and keeps the earlier one readable.
Act on this result
Nothing filed here edits the run or the grade. A challenge opens a review, and a review can produce a new run or a re-grade — which becomes the current result and leaves this one in the history.
You run the same kind of test against your own setup and get the same behaviour.
Agree →Yours behaves differently. Tell us what you got, with a screenshot if you have one.
Disagree →Something here is wrong — a reference value, a transcription, a grade.
Report an issue →Tell us what changed and we schedule a rerun of the failing test case. The old result stays as history.
Vendor notice →The same record is available as structured data through the AI Demos MCP server, with the counts, the coverage and every per-test-case reason carried as fields.
Verify the proof files
These files support this result. Open a file to inspect the original evidence.
File fingerprints (SHA-256)
A fingerprint identifies the exact file used for this result.