Research

Paper

AI LLM March 19, 2026

DaPT: A Dual-Path Framework for Multilingual Multi-hop Question Answering

Authors

Yilin Wang, Yuchun Fan, Jiaoyang Li, Ziming Zhu, Yongyu Mu, Qiaozhi He, Tong Xiao, Jingbo Zhu

Abstract

Retrieval-augmented generation (RAG) systems have made significant progress in solving complex multi-hop question answering (QA) tasks in the English scenario. However, RAG systems inevitably face the application scenario of retrieving across multilingual corpora and queries, leaving several open challenges. The first one involves the absence of benchmarks that assess RAG systems' capabilities under the multilingual multi-hop (MM-hop) QA setting. The second centers on the overreliance on LLMs' strong semantic understanding in English, which diminishes effectiveness in multilingual scenarios. To address these challenges, we first construct multilingual multi-hop QA benchmarks by translating English-only benchmarks into five languages, and then we propose DaPT, a novel multilingual RAG framework. DaPT generates sub-question graphs in parallel for both the source-language query and its English translation counterpart, then merges them before employing a bilingual retrieval-and-answer strategy to sequentially solve sub-questions. Our experimental results demonstrate that advanced RAG systems suffer from a significant performance imbalance in multilingual scenarios. Furthermore, our proposed method consistently yields more accurate and concise answers compared to the baselines, significantly enhancing RAG performance on this task. For instance, on the most challenging MuSiQue benchmark, DaPT achieves a relative improvement of 18.3\% in average EM score over the strongest baseline.

Metadata

arXiv ID: 2603.19097
Provider: ARXIV
Primary Category: cs.CL
Published: 2026-03-19
Fetched: 2026-03-20 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.19097v1</id>\n    <title>DaPT: A Dual-Path Framework for Multilingual Multi-hop Question Answering</title>\n    <updated>2026-03-19T16:23:04Z</updated>\n    <link href='https://arxiv.org/abs/2603.19097v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.19097v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Retrieval-augmented generation (RAG) systems have made significant progress in solving complex multi-hop question answering (QA) tasks in the English scenario. However, RAG systems inevitably face the application scenario of retrieving across multilingual corpora and queries, leaving several open challenges. The first one involves the absence of benchmarks that assess RAG systems' capabilities under the multilingual multi-hop (MM-hop) QA setting. The second centers on the overreliance on LLMs' strong semantic understanding in English, which diminishes effectiveness in multilingual scenarios. To address these challenges, we first construct multilingual multi-hop QA benchmarks by translating English-only benchmarks into five languages, and then we propose DaPT, a novel multilingual RAG framework. DaPT generates sub-question graphs in parallel for both the source-language query and its English translation counterpart, then merges them before employing a bilingual retrieval-and-answer strategy to sequentially solve sub-questions. Our experimental results demonstrate that advanced RAG systems suffer from a significant performance imbalance in multilingual scenarios. Furthermore, our proposed method consistently yields more accurate and concise answers compared to the baselines, significantly enhancing RAG performance on this task. For instance, on the most challenging MuSiQue benchmark, DaPT achieves a relative improvement of 18.3\\% in average EM score over the strongest baseline.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CL'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <published>2026-03-19T16:23:04Z</published>\n    <arxiv:comment>Accepted by ICASSP 2026</arxiv:comment>\n    <arxiv:primary_category term='cs.CL'/>\n    <author>\n      <name>Yilin Wang</name>\n    </author>\n    <author>\n      <name>Yuchun Fan</name>\n    </author>\n    <author>\n      <name>Jiaoyang Li</name>\n    </author>\n    <author>\n      <name>Ziming Zhu</name>\n    </author>\n    <author>\n      <name>Yongyu Mu</name>\n    </author>\n    <author>\n      <name>Qiaozhi He</name>\n    </author>\n    <author>\n      <name>Tong Xiao</name>\n    </author>\n    <author>\n      <name>Jingbo Zhu</name>\n    </author>\n  </entry>"
}