Transforming Questions and Documents for Semantically Aligned Retrieval-Augmented Generation
By: Seokgi Lee
Potential Business Impact:
Answers hard questions by breaking them down.
We introduce a novel retrieval-augmented generation (RAG) framework tailored for multihop question answering. First, our system uses large language model (LLM) to decompose complex multihop questions into a sequence of single-hop subquestions that guide document retrieval. This decomposition mitigates the ambiguity inherent in multi-hop queries by clearly targeting distinct knowledge facets. Second, instead of embedding raw or chunked documents directly, we generate answerable questions from each document chunk using Qwen3-8B, embed these generated questions, and retrieve relevant chunks via question-question embedding similarity. During inference, the retrieved chunks are then fed along with the original question into the RAG pipeline. We evaluate on three multihop question datasets (MuSiQue, 2WikiMultiHopQa, HotpotQA) from LongBench. Our method improves RAG performacne compared to baseline systems. Our contributions highlight the benefits of using answerable-question embeddings for RAG, and the effectiveness of LLM-based query decomposition for multihop scenarios.
Similar Papers
Knowledge Compression via Question Generation: Enhancing Multihop Document Retrieval without Fine-tuning
Information Retrieval
Helps computers find answers by asking questions.
Enhancing Document-Level Question Answering via Multi-Hop Retrieval-Augmented Generation with LLaMA 3
Computation and Language
Answers hard questions from long texts better.
Optimizing Multi-Hop Document Retrieval Through Intermediate Representations
Computation and Language
Helps AI answer harder questions faster.