Retrieve, Summarize, Plan: Advancing Multi-hop Question Answering with an Iterative Approach
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912209726078976 |
|---|---|
| author | Jiang, Zhouyu Sun, Mengshu Liang, Lei Zhang, Zhiqiang |
| author_facet | Jiang, Zhouyu Sun, Mengshu Liang, Lei Zhang, Zhiqiang |
| contents | Multi-hop question answering is a challenging task with distinct industrial relevance, and Retrieval-Augmented Generation (RAG) methods based on large language models (LLMs) have become a popular approach to tackle this task. Owing to the potential inability to retrieve all necessary information in a single iteration, a series of iterative RAG methods has been recently developed, showing significant performance improvements. However, existing methods still face two critical challenges: context overload resulting from multiple rounds of retrieval, and over-planning and repetitive planning due to the lack of a recorded retrieval trajectory. In this paper, we propose a novel iterative RAG method called ReSP, equipped with a dual-function summarizer. This summarizer compresses information from retrieved documents, targeting both the overarching question and the current sub-question concurrently. Experimental results on the multi-hop question-answering datasets HotpotQA and 2WikiMultihopQA demonstrate that our method significantly outperforms the state-of-the-art, and exhibits excellent robustness concerning context length. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_13101 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Retrieve, Summarize, Plan: Advancing Multi-hop Question Answering with an Iterative Approach Jiang, Zhouyu Sun, Mengshu Liang, Lei Zhang, Zhiqiang Computation and Language Artificial Intelligence Multi-hop question answering is a challenging task with distinct industrial relevance, and Retrieval-Augmented Generation (RAG) methods based on large language models (LLMs) have become a popular approach to tackle this task. Owing to the potential inability to retrieve all necessary information in a single iteration, a series of iterative RAG methods has been recently developed, showing significant performance improvements. However, existing methods still face two critical challenges: context overload resulting from multiple rounds of retrieval, and over-planning and repetitive planning due to the lack of a recorded retrieval trajectory. In this paper, we propose a novel iterative RAG method called ReSP, equipped with a dual-function summarizer. This summarizer compresses information from retrieved documents, targeting both the overarching question and the current sub-question concurrently. Experimental results on the multi-hop question-answering datasets HotpotQA and 2WikiMultihopQA demonstrate that our method significantly outperforms the state-of-the-art, and exhibits excellent robustness concerning context length. |
| title | Retrieve, Summarize, Plan: Advancing Multi-hop Question Answering with an Iterative Approach |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2407.13101 |