XProvence: Zero-Cost Multilingual Context Pruning for Retrieval-Augmented Generation
Preprint
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.Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- Context Embeddings for Efficient Answer Generation in Retrieval-Augmented Generation2025
- MedExpQA: Multilingual benchmarking of Large Language Models for Medical Question AnsweringArtificial Intelligence in Medicine, 2024
- CompAct: Compressing Retrieved Documents Actively for Question Answering2024
- M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation2024
- MIRACL: A Multilingual Retrieval Dataset Covering 18 Diverse LanguagesTransactions of the Association for Computational Linguistics, 2023
- Is ChatGPT Good at Search? Investigating Large Language Models as Re-Ranking Agents2023
- MKQA: A Linguistically Diverse Benchmark for Multilingual Open Domain Question AnsweringTransactions of the Association for Computational Linguistics, 2021
- TyDi QA: A Benchmark for Information-Seeking Question Answering in Typologically Diverse LanguagesTransactions of the Association for Computational Linguistics, 2020
- Unsupervised Cross-lingual Representation Learning at Scale2020
- On the Cross-lingual Transferability of Monolingual Representations2020