CVApr 29

When Good OCR Is Not Enough: Benchmarking OCR Robustness for Retrieval-Augmented Generation

arXiv:2605.0091190.9h-index: 3Has Code
Predicted impact top 14% in CV · last 90 daysOriginality Incremental advance
AI Analysis

For industrial RAG systems, this benchmark reveals a critical gap between OCR metrics and actual retrieval/generation performance.

Existing OCR benchmarks fail to predict downstream RAG performance on industrial documents. The authors introduce a benchmark covering 11 challenging document types and show that high OCR accuracy does not guarantee strong RAG results due to structural and semantic errors.

Industrial Retrieval-Augmented Generation (RAG) systems depend on optical character recognition (OCR) to transform visual documents into text. Existing OCR benchmarks rely on character-level metrics, which inadequately measure downstream RAG effectiveness under real-world conditions. We introduce an OCR benchmark for industrial RAG systems covering 11 challenging document types, including extreme layouts, high-resolution pages, complex or watermarked backgrounds, historical documents with non-standard reading orders, visually decorated text, and documents containing tables and mathematical formulas. Evaluating recent SOTA OCR models under a controlled OCR-first RAG pipeline shows clear performance degradation on realistic industrial documents despite strong conventional benchmark scores. We find that high OCR accuracy does not necessarily translate into strong downstream RAG performance: structural and semantic errors can cause substantial retrieval failures even when WER/CER remains low. Further analysis shows that this mismatch is category-dependent, arises through both retrieval-side and downstream generation-side failures, and remains stable across representative OCR-first pipeline choices. The benchmark is publicly available at https://github.com/Qihoo360/InduOCRBench.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes