English

XProvence: Zero-Cost Multilingual Context Pruning for Retrieval-Augmented Generation

Information Retrieval 2026-01-28 v1 Computation and Language

Abstract

This paper introduces XProvence, a multilingual zero-cost context pruning model for retrieval-augmented generation (RAG), trained on 16 languages and supporting 100+ languages through effective cross-lingual transfer. Motivated by the growing use of RAG systems across diverse languages, we explore several strategies to generalize the Provence framework-which first integrated efficient zero-cost context pruning directly into the re-ranking model-beyond English. Across four multilingual question answering benchmarks, we show how XProvence can prune RAG contexts with minimal-to-no performance degradation and outperforms strong baselines. Our model is available at https://huggingface.co/naver/xprovence-reranker-bgem3-v2.

Cite

@article{arxiv.2601.18886,
  title  = {XProvence: Zero-Cost Multilingual Context Pruning for Retrieval-Augmented Generation},
  author = {Youssef Mohamed and Mohamed Elhoseiny and Thibault Formal and Nadezhda Chirkova},
  journal= {arXiv preprint arXiv:2601.18886},
  year   = {2026}
}

Comments

Accepted to ECIR 2026

R2 v1 2026-07-01T09:21:04.999Z