References
Data references and terminology definitions for document AI. Each page is source-verified and designed to be cited.
Statistics
Receipt v1.1: docTR vs Docling on fixed SROIE and CORD receipts; normalized text distance, postprocessed field F1, batch throughput, GPU cost.
Receipt v1.1: docTR vs Surya2 on 361 SROIE and 100 CORD receipts; text-distance diagnostics, postprocessed field F1, batch timing, GPU cost.
Receipt v1.1: docTR leads SROIE LLM field F1 and batch throughput; EasyOCR leads SROIE regex F1 and CORD batch throughput. Methods and sources.
Receipt benchmark: available RTX 4090 GPU estimates span $0.0479–$1.0609 per 1,000 SROIE pages at historical $0.76/hr. Timing limits and CPU scope explained.
Receipt OCR request p50 spans 108.7–2668.0 ms on SROIE. Eight retained CPU/GPU engine runs; p95, accepted wall throughput and timing limits explained.
Receipt v1.1 compares PaddleOCR and EasyOCR on SROIE and CORD: normalized text distance, regex and LLM field F1, batch throughput, GPU cost.
First-party benchmark: LLM postprocessing beats regex for receipt field extraction, field F1 for 8 OCR models on SROIE + CORD, methodology included.
Receipt v1.1 compares three VLMs on SROIE and CORD field scores, request latency, and historical GPU cost; Unlimited-OCR SROIE wall and cost missing.
Receipt v1.1: Tesseract vs PaddleOCR on 361 SROIE and 100 CORD receipts; normalized text distance, field F1, missing wall time, CPU cost scope.
First-party benchmark: 8 open-source OCR engines vs document-parsing VLMs on 361 SROIE + 100 CORD receipts — CER, field F1, latency, cost. Methodology included.
Receipt benchmark: normalized SROIE CER and field F1 measure different targets. CORD whole-page CER is not applicable. Scoring and pipeline limits explained.
First-party benchmark of 5 open-source OCR models on 461 receipts (SROIE + CORD): CER, WER, field F1, latency, cost per 1,000 pages.
Exception rates in document automation: 9% best-in-class vs 22% average, root causes, review costs, and threshold tradeoffs with all sources cited.
Independent benchmarks on document processing cost per record: manual $8-16, template $2-5, AI from $0.02/page. By document type, size, geography.
Handwriting OCR accuracy data: IAM CER 1.2-1.7%, Arabic 8.9-16%, historical 71-82%. Broken down by script, model & document type. 15 citable sources.
Benchmarks on invoice processing time: 12.5 min manual touch vs 9.2-day cycle, by method, complexity, stage — with full sources for citation.
Independent benchmarks on data entry automation ROI: 3-12 month payback, 60-80% savings, breakeven volume, plus a formula to model your return.
Independent benchmarks on OCR accuracy by document type: 99.2-99.8% digital PDFs, 98-99% scans, 46-95% handwriting, table TEDS. 15 citable sources.
Receipt OCR accuracy by condition and field: 90%+ clean scans, 55-73% faded thermal, 67-69% phone photos. Totals 95-99%, merchants 73-88%. 14 sources.
Independent benchmarks on manual data entry error rates: field-level 1-4%, record-level up to 40%, by document type, industry, and role.
Definitions
Data capture turns paper, scans, PDFs, and photos into machine-readable data. How it works, its types, and misconceptions.
ICR is machine-learning character recognition for hand-printed text. How it evolved from OCR, the ICR vs OCR vs IWR differences, and where it's used.
Key Information Extraction (KIE) pulls fields like invoice numbers and dates from documents into structured data. How it works, boundaries, misconceptions.
A searchable PDF is a scan with an invisible OCR text layer — you can select, copy, and Ctrl+F the text. How it works, PDF/A rules, and common traps.
Field-level accuracy counts correctly extracted fields; character-level counts correctly read characters. 99% character accuracy is not 99% accurate data.
IDP is AI software that captures, classifies, extracts, and validates data from documents. How it works, evolution from OCR to LLMs, and misconceptions.
STP rate measures the share of documents processed with zero human touch — industry avg 32%, best-in-class 49%, AI 60-80%. Formula and misconceptions.
OCR converts images of printed, handwritten, or typed text into machine-readable data. How it works, the 4 types of OCR, its history, and use cases.