# Hybrid Retrieval-Augmented Generation for Robust Multilingual Document Question Answering

Source: [Hybrid Retrieval-Augmented Generation for Robust Multilingual Document Question Answering](<https://www.sota2.com/research/paper/6940e7ca456146936284fa82>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-12-17T05:04:46.379Z

Authors: Anthony Mudet, Souhail Bakkali

Published: 2025-12-14T13:57:05.000Z

## Abstract

Large-scale digitization initiatives have unlocked massive collections of historical newspapers, yet effective computational access remains hindered by OCR corruption, multilingual orthographic variation, and temporal language drift. We develop and evaluate a multilingual Retrieval-Augmented Generation pipeline specifically designed for question answering on noisy historical documents. Our approach integrates: (i) semantic query expansion and multi-query fusion using Reciprocal Rank Fusion to improve retrieval robustness against vocabulary mismatch; (ii) a carefully engineered generation prompt that enforces strict grounding in retrieved evidence and explicit abstention when evidence is insufficient; and (iii) a modular architecture enabling systematic component evaluation. We conduct comprehensive ablation studies on Named Entity Recognition and embedding model selection, demonstrating the importance of syntactic coherence in entity extraction and balanced performance-efficiency trade-offs in dense retrieval. Our end-to-end evaluation framework shows that the pipeline generates faithful answers for well-supported queries while correctly abstaining from unanswerable questions. The hybrid retrieval strategy improves recall stability, particularly benefiting from RRF's ability to smooth performance variance across query formulations. We release our code and configurations at https://anonymous.4open.science/r/RAGs-C5AE/, providing a reproducible foundation for robust historical document question answering.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2512.12694>)

[PDF](<https://arxiv.org/pdf/2512.12694.pdf>)