Leveraging Inter-Chunk Interactions for Enhanced Retrieval in Large Language Model-Based Question Answering
This addresses retrieval challenges in complex question-answering tasks, but it is incremental as it builds on existing retrieval paradigms.
The paper tackled the problem of ambiguous references and lack of context in retrieval for large language model-based question answering by proposing a new framework that leverages inter-chunk interactions, resulting in outperforming strong baselines across four datasets.
Retrieving external knowledge and prompting large language models with relevant information is an effective paradigm to enhance the performance of question-answering tasks. Previous research typically handles paragraphs from external documents in isolation, resulting in a lack of context and ambiguous references, particularly in multi-document and complex tasks. To overcome these challenges, we propose a new retrieval framework IIER, that leverages Inter-chunk Interactions to Enhance Retrieval. This framework captures the internal connections between document chunks by considering three types of interactions: structural, keyword, and semantic. We then construct a unified Chunk-Interaction Graph to represent all external documents comprehensively. Additionally, we design a graph-based evidence chain retriever that utilizes previous paths and chunk interactions to guide the retrieval process. It identifies multiple seed nodes based on the target question and iteratively searches for relevant chunks to gather supporting evidence. This retrieval process refines the context and reasoning chain, aiding the large language model in reasoning and answer generation. Extensive experiments demonstrate that IIER outperforms strong baselines across four datasets, highlighting its effectiveness in improving retrieval and reasoning capabilities.