Retrieval-augmented generation
MiniCPM (OCR)
Superseded — cited as a baseline and beaten by newer methods
0 papers critique it · 2 beat it on benchmarks
Beaten on benchmarks
Head-to-head results where a newer method reports beating MiniCPM (OCR). Values are copied from the source paper's tables — verify against the cited paper.
VisRAG-Ret beats MiniCPM (OCR)
69.17 vs 47.96
Average MRR@10 · [Out-of-domain: Models Fine-tuned on Synthetic Data]
VisRAG: Vision-based Retrieval-augmented Generation on Multi-modality DocumentsVisRAG-Gen Page Concatenation beats MiniCPM (OCR)
36.16 vs 27.68
Average accuracy top-1 · [Text-based Generation with MiniCPM (OCR)]
VisRAG: Vision-based Retrieval-augmented Generation on Multi-modality DocumentsVisRAG-Ret beats MiniCPM (OCR)
77.91 vs 74.78
Average MRR@10 · [MLLMs as End-to-End Representers]
Roles of MLLMs in Visually Rich Document Retrieval for RAG: A Survey
What to use instead
Recent methods in the same sub-problem, not yet superseded in the knowledge base — arXiv benchmark leaders, not vetted production recommendations.
- Apr 7, 2026
- Graph-to-Frame RAGGraph-to-Frame RAG: Visual-Space Knowledge Fusion for Training-Free and Auditable Video ReasoningApr 6, 2026
- Apr 4, 2026
- AutoThinkRAGAutothinkRAG: Complexity-Aware Control of Retrieval-Augmented Reasoning for Image-Text InteractionMar 17, 2026
- Feb 27, 2026
- VimRAGVimRAG: Navigating Massive Visual Context in Retrieval-Augmented Generation via Multimodal Memory GraphFeb 13, 2026
- Feb 5, 2026
- Feb 1, 2026
- Oct 8, 2025