Cite Datalab as inspiration, prominent viewer link (#12)

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

Author: davanstrien

CLAUDE.md

@@ -4,6 +4,8 @@ OCR model evaluation toolkit. Answers: **"Which OCR model works best for MY docu
 
 Rankings change by document type — the best model for manuscript cards is different from the best for printed books or historical texts. This tool creates per-collection leaderboards using pairwise VLM-as-judge comparisons, so users can find what works for their specific documents.
 
+Inspired by [Datalab's Benchmarks + Evals](https://www.datalab.to/blog/datalab-benchmarks-evals) — pairwise VLM-as-judge with Bradley-Terry scoring per document class — but as an open-source, Hub-native tool anyone can run on their own collections.
+
 **Pipeline**: `run` (launch OCR models via HF Jobs) → `judge` (pairwise VLM comparison → Bradley-Terry ELO) → `view` (leaderboard + human validation). Everything lives on the Hugging Face Hub — no local GPU needed.
 
 ## Architecture

README.md

@@ -8,7 +8,7 @@ The underlying OCR model inference uv scripts are available at [uv-scripts/ocr](
 
 ## Why?
 
-Generic OCR benchmarks tell you which model wins _on average_. But if you're digitising 18th-century encyclopaedias, that average doesn't help — the best model for your documents might be the worst on someone else's.
+Generic OCR benchmarks tell you which model wins _on average_. But if you're digitising 18th-century encyclopaedias, that average doesn't help — the best model for your documents might be the worst on someone else's. Inspired by [Datalab's Benchmarks + Evals](https://www.datalab.to/blog/datalab-benchmarks-evals) approach — pairwise VLM-as-judge with Bradley-Terry scoring on your own documents — ocr-bench brings this idea to the Hugging Face Hub as an open-source, self-serve tool.
 
 ocr-bench lets you run the same set of OCR models on a sample of _your_ collection, then uses a vision-language model to judge which produces the best transcription for each document. The result is a leaderboard specific to your data.
 
@@ -24,6 +24,8 @@ Rankings can flip completely between collections.
 
 ![ELO vs Parameter Count — smaller models can win on the right documents](assets/elo-scatter.png)
 
+**[Try the live viewer](https://huggingface.co/spaces/davanstrien/ocr-bench-britannica-results-qwen35-viewer)** — browse the Britannica 1771 leaderboard, compare OCR outputs side-by-side, and vote on quality yourself.
+
 ## Hub-native by design
 
 The entire evaluation loop lives on the Hugging Face Hub: