Evaluating OCR programs that convert PDFs or doc photos into Markdown is way extra advanced than it seems. In contrast to plain textual content OCR, OCR-to-Markdown requires fashions to recuperate content material, format, studying order, and illustration selections concurrently. As we speak’s benchmarks try to attain this with a mixture of string matching, heuristic alignment, and format-specific guidelines—however in apply, these approaches routinely misclassify appropriate outputs as failures.
This put up outlines why OCR-to-Markdown analysis is inherently underspecified, examines frequent analysis methods and their failure modes, highlights concrete points noticed in two broadly used benchmarks, and explains why LLM-as-judge is presently essentially the most sensible approach to consider these programs—regardless of its imperfections .
Why OCR-to-Markdown Is Onerous to Consider
At its core, OCR-to-Markdown doesn’t have a single appropriate output.
A number of outputs will be equally legitimate:
- Multi-column layouts will be linearized in several studying orders.
- Equations will be represented utilizing LaTeX, Unicode, HTML, or hybrids.
- Headers, footers, watermarks, and marginal textual content could or is probably not thought-about “content material” relying on process intent.
- Spacing, punctuation, and Unicode normalization usually differ with out affecting which means.
From a human or downstream-system perspective, these outputs are equal. From a benchmark’s perspective, they usually should not.
Frequent Analysis Strategies and Their Limitations
1. String-Primarily based Metrics (Edit Distance, Actual Match)
Most OCR-to-Markdown benchmarks depend on normalized string comparability or edit distance.
Limitations
- Markdown is handled as a flat character sequence, ignoring construction.
- Minor formatting variations produce giant penalties.
- Structurally incorrect outputs can rating properly if textual content overlaps.
- Scores correlate poorly with human judgment.
These metrics reward formatting compliance moderately than correctness.
2. Order-Delicate Block Matching
Some benchmarks phase paperwork into blocks and rating ordering and proximity.
Limitations
- Legitimate various studying orders (e.g., multi-column paperwork) are penalized.
- Small footer or marginal textual content can break strict ordering constraints.
- Matching heuristics degrade quickly as format complexity will increase.
Appropriate content material is usually marked incorrect as a consequence of ordering assumptions.
3. Equation Matching by way of LaTeX Normalization
Math-heavy benchmarks usually anticipate equations to be rendered as full LaTeX.
Limitations
- Unicode or partially rendered equations are penalized.
- Equal LaTeX expressions utilizing totally different macros fail to match.
- Blended LaTeX/Markdown/HTML representations should not dealt with.
- Rendering-correct equations nonetheless fail string-level checks.
This conflates illustration selection with mathematical correctness.
4. Format-Particular Assumptions
Benchmarks implicitly encode a most popular output model.
Limitations
- HTML tags (e.g.,
) trigger matching failures. - Unicode symbols (e.g.,
km²) are penalized towards LaTeX equivalents. - Spacing and punctuation inconsistencies in floor reality amplify errors.
Fashions aligned to benchmark formatting outperform extra common OCR programs.
Points Noticed in Current Benchmarks
Benchmark A: olmOCRBench
Guide inspection reveals that a number of subsets embed implicit content material omission guidelines:
- Headers, footers, and watermarks which might be visibly current in paperwork are explicitly marked as absent in floor reality.
- Fashions educated to extract all seen textual content are penalized for being appropriate.
- These subsets successfully consider selective suppression, not OCR high quality.
Moreover:
- Math-heavy subsets fail when equations should not totally normalized LaTeX.
- Appropriate predictions are penalized as a consequence of illustration variations.
In consequence, scores strongly rely upon whether or not a mannequin’s output philosophy matches the benchmark’s hidden assumptions.
Instance 1
For the above picture, Nanonets-OCR2 accurately predicts the watermark to the correct facet of the picture, however within the floor reality annotation penalizes the mannequin for predicting it accurately.
{
"pdf": "headers_footers/ef5e1f5960b9f865c8257f9ce4ff152a13a2559c_page_26.pdf",
"web page": 1,
"id": "ef5e1f5960b9f865c8257f9ce4ff152a13a2559c_page_26.pdf_manual_01",
"sort": "absent",
"textual content": "Doc tu00e9lu00e9chargu00e9 depuis www.cairn.information - Universitu00e9 de Marne-la-Vallu00e9e - - 193.50.159.70 - 20/03/2014 09h07. u00a9 S.A.C.", "case_sensitive": false, "max_diffs": 3, "checked": "verified", "first_n": null, "last_n": null, "url": ""}
Sort absent implies that within the prediction information, that textual content shouldn’t be current.
Instance 2
The benchmark additionally doesn’t take into account texts which might be current within the doc footer.
Instance on this doc, the Alcoholics Namelessu00ae and www.aa.org shouldn’t be current within the doc in accordance with the ground-truth, which is inaccurate
{
"pdf": "headers_footers/3754542bf828b42b268defe21db8526945928834_page_4.pdf",
"web page": 1,
"id": "3754542bf828b42b268defe21db8526945928834_page_4_header_00",
"sort": "absent",
"max_diffs": 0,
"checked": "verified",
"url": "",
"textual content": "Alcoholics Namelessu00ae",
"case_sensitive": false, "first_n": null, "last_n": null
}
{
"pdf": "headers_footers/3754542bf828b42b268defe21db8526945928834_page_4.pdf",
"web page": 1,
"id": "3754542bf828b42b268defe21db8526945928834_page_4_header_01",
"sort": "absent",
"max_diffs": 0,
"checked": "verified",
"url": "",
"textual content": "www.aa.org",
"case_sensitive": false, "first_n": null, "last_n": null}
Benchmark B: OmniDocBench
OmniDocBench reveals related points, however extra broadly:
- Equation analysis depends on strict LaTeX string equivalence.
- Semantically equivalent equations fail as a consequence of macro, spacing, or image variations.
- Quite a few ground-truth annotation errors had been noticed (lacking tokens, malformed math, incorrect spacing).
- Unicode normalization and spacing variations systematically cut back scores.
- Prediction choice heuristics can fail even when the proper reply is totally current.
In lots of circumstances, low scores replicate benchmark artifacts, not mannequin errors.
Instance 1
Within the instance above, the Nanonets-OCR2-3B predicts 5 g silica + 3 g Al$_2$O$_3$ however the floor reality expects as $ 5g \mathrm{\ s i l i c a}+3g \mathrm{\ A l}*{2} \mathrm{O*{3}} $ . This flags the mannequin prediction as incorrect, even when each are appropriate.
Full Floor Fact and Prediction, and the take a look at case shared under:
'pred': 'The collected eluant was concentrated by rotary evaporator to 1 ml. The extracts had been lastly handed via a ultimate column full of 5 g silica + 3 g Al$_2$O$_3$ to take away any co-extractive compounds which will trigger instrumental interferences durin the evaluation. The extract was eluted with 120 ml of DCM:n-hexane (1:1), the primary 18 ml of eluent was discarded and the remainder had been collected, which comprises the analytes of curiosity. The extract was exchanged into n-hexane, concentrated to 1 ml to which 1 μg/ml of inner normal was added.'
'gt': 'The collected eluant was concentrated by rotary evaporator to 1 ml .The extracts had been lastly handed via a ultimate column full of $ 5g \mathrm{\ s i l i c a}+3g \mathrm{\ A l}*{2} \mathrm{O*{3}} $ to take away any co-extractive compounds which will trigger instrumental
interferences in the course of the evaluation. The extract was eluted with 120 ml of DCM:n-hexane (1:1), the primary 18 ml of eluent was discarded and the remainder had been collected, which comprises the analytes of curiosity. The extract was exchanged into n - hexane, concentrated to 1 ml to which $ \mu\mathrm{g / ml} $ of inner normal was added.'Instance 2
We discovered considerably extra incorrect annotations with OmniDocBench
Within the ground-truth annotation 1 is lacking in 1 ml .
'textual content': 'The collected eluant was concentrated by rotary evaporator to 1 ml .The extracts had been lastly handed via a ultimate column full of $ 5g \mathrm{\ s i l i c a}+3g \mathrm{\ A l}*{2} \mathrm{O*{3}} $ to take away any co-extractive compounds which will trigger instrumental interferences in the course of the evaluation. The extract was eluted with 120 ml of DCM:n-hexane (1:1), the primary 18 ml of eluent was discarded and the remainder had been collected, which comprises the analytes of curiosity. The extract was exchanged into n - hexane, concentrated to 1 ml to which $ \mu\mathrm{g / ml} $ of inner normal was added.'
