Reproducibility, Replicability, And Insights Into Visual Document Retrieval With Late Interaction

Abstract

Visual Document Retrieval (VDR) is an emerging research area that focuses on encoding and retrieving document images directly, bypassing the dependence on Optical Character Recognition (OCR) for document search. A recent advance in VDR was introduced by ColPali, which significantly improved retrieval effectiveness through a late interaction mechanism. ColPali's approach demonstrated substantial performance gains over existing baselines that do not use late interaction on an established benchmark. In this study, we investigate the reproducibility and replicability of VDR methods with and without late interaction mechanisms by systematically evaluating their performance across multiple pre-trained vision-language models. Our findings confirm that late interaction yields considerable improvements in retrieval effectiveness; however, it also introduces computational inefficiencies during inference. Additionally, we examine the adaptability of VDR models to textual inputs and assess their r

Stats

• citations1
• S2 citations—
• github stars0
• HF likes0
• heat score2.26
• arxiv keyqiao2025reproducibility

Related papers

• Modernvbert: Towards Smaller Visual Document Retrievers (2025)0.00
• Colpali: Efficient Document Retrieval With Vision Language Models (2024)0.00
• SERVAL: Surprisingly Effective Zero-shot Visual Document Retrieval Powered By Large Vision And Language Models (2025)0.00
• Visual Late Chunking: An Empirical Study Of Contextual Chunking For Efficient Visual Document Retrieval (2026)0.00
• Unlocking Multimodal Document Intelligence: From Current Triumphs To Future Frontiers Of Visual Document Retrieval (2026)0.00
• Colbertv2: Effective And Efficient Retrieval Via Lightweight Late Interaction (2021)17.46
• Nemotron Colembed V2: Top-performing Late Interaction Embedding Models For Visual Document Retrieval (2026)0.00
• Docpruner: A Storage-efficient Framework For Multi-vector Visual Document Retrieval Via Adaptive Patch-level Embedding Pruning (2025)0.00