--- pretty_name: AURA OCR Benchmark — Real5-OmniDocBench Illumination (2026-11-20) task_categories: - image-to-text tags: - ocr - document-layout-analysis - omnidocbench - benchmark size_categories: - 1K/ metadata.txt # per-file timing page_001/ page.png # the page as sent to the endpoint layout.json # regions: label, category, score, bbox, reading order, crop file layout_vis.png # boxes drawn on the page crop______.png vllm_benchmark/ # Stage 2 output VLLM_Report_51186.pdf # benchmark report vllm_benchmark_summary.json # metrics for the report aura-ocr_layout_enabled_formula/ omnidoc_benchmark_summary.{json,txt} # accuracy summary errors.txt # files that failed / markdown.md # final page Markdown metadata.txt # per-page latency and token details crops/ # figure crops referenced by markdown.md (only on pages with figures) aura_illum_layout.log # Stage 1 console log aura_illum_ocr.log # Stage 2 console log ``` Each `layout.json` region looks like this: ```json {"id": 1, "label": "paragraph_title", "cls_id": 17, "order": 0, "category": "text", "score": 0.8782, "bbox": [1586, 302, 2248, 438], "crop_filename": "crop_001_text_1586_302_2248_438.png"} ``` ## Known issues - **One page was not scored.** `color_textbook_…人教新起点英语5A电子课本_page_004` contains only illustrations with blacked-out labels. The layout endpoint found 0 regions, so the OCR stage logged it in `errors.txt` instead of saving an empty page. - **Transient errors during OCR.** A few HTTP 429 (rate limit) responses and dropped connections happened; all succeeded on retry. One region returned no text after 3 retries and was saved as blank. - **The layout run was resumed once.** The layout timing (3.84 s per file) comes from the resumed run, which covered 1,164 files; the first 191 files were skipped. That run also includes a few slow files caused by a short DNS outage, which add about 0.04 s per file.