paper-with-me

홈 › Papers

IndexRAG: Bridging Facts for Cross-Document Reasoning at Index Time

2026-03-17 · Zhenghua Bao, Yi Shi arxiv

Multi-hop question answering (QA) requires reasoning across multiple documents, yet existing retrieval-augmented generation (RAG) approaches address this either through graph-based methods requiring additional online processing or iterative multi-step reasoning. We present IndexRAG, a novel approach that shifts cross-document reasoning from online inference to offline indexing. IndexRAG identifies bridge entities shared across documents and generates bridging facts as independently retrievable units, requiring no additional training or fine-tuning. Experiments on three widely-used multi-hop QA benchmarks (HotpotQA, 2WikiMultiHopQA, MuSiQue) show that IndexRAG improves F1 over Naive RAG by 4.6 points on average, while requiring only single-pass retrieval and a single LLM call at inference time. When combined with IRCoT, IndexRAG outperforms all graph-based baselines on average, including HippoRAG and FastGraphRAG, while relying solely on flat retrieval. Our code will be released upon acceptance.

📄 PDF Abstract BibTeX arXiv:2603.16415

Code (0)

등록된 구현이 없습니다.

Tasks

Multi-hop Question Answering

Similar Papers 제목 키워드 기반

CodRED: A Cross-Document Relation Extraction Dataset for Acquiring Knowledge in the Wild

2021-11-01 · EMNLP 2021 11 · Yuan YAO, Jiaju Du, Yankai Lin, Peng Li 외

Existing relation extraction (RE) methods typically focus on extracting relational facts between entity pairs within single sentences or documents. However, a large quantity of relational facts in knowledge bases can onl…

RelationRelation Extraction

Semantic Bridge: Universal Multi-Hop Question Generation via AMR-Driven Graph Synthesis

2025-08-06 · Linqing Chen, Hanmeng Zhong, Wentao Wu, Weilei Wang arxiv

Large language model (LLM) training faces a critical bottleneck: the scarcity of high-quality, reasoning-intensive question-answer pairs, especially from sparse, domain-specific sources like PubMed papers or legal docume…

Question Generation

Do Language Models Reason Across Languages?

2026-01-10 · Yan Meng, Wafaa Mohammed, Christof Monz arxiv

The real-world information sources are inherently multilingual, which naturally raises a question about whether language models can synthesize information across languages. In this paper, we introduce a simple two-hop qu…

Question Answering

Reading, Not Thinking: Understanding and Bridging the Modality Gap When Text Becomes Pixels in Multimodal LLMs

2026-03-10 · Kaiser Sun, Xiaochuang Yuan, Hongjun Liu, Chen Zhao 외 arxiv

Multimodal large language models (MLLMs) can process text presented as images, yet they often perform worse than when the same content is provided as textual tokens. We systematically diagnose this "modality gap" by eval…

BRIEF: Bridging Retrieval and Inference for Multi-hop Reasoning via Compression

2024-10-20 · Yuankai Li, Jia-Chen Gu, Di wu, Kai-Wei Chang 외

Retrieval-augmented generation (RAG) can supplement large language models (LLMs) by integrating external knowledge. However, as the number of retrieved documents increases, the input length to LLMs grows linearly, causin…

In-Context LearningLong-Context UnderstandingOpen-Domain Question AnsweringQuestion Answering+3