Generate-then-Ground in Retrieval-Augmented Generation for Multi-hop Question Answering

Zhengliang Shi, Shuo Zhang, Weiwei Sun, Shen Gao, Pengjie Ren, Zhumin Chen, Zhaochun Ren


Abstract
Multi-Hop Question Answering (MHQA) task presents a significant challenge for large language models (LLMs) due to the intensive knowledge required. Current solutions, like Retrieval-Augmented Generation, typically retrieve potential documents from an external corpus to read an answer. However, the performance of this retrieve-then-read paradigm is constrained by the retriever and the inevitable noise in the retrieved documents. To mitigate these challenges, we introduce a novel generate-then-ground (GenGround) framework, synergizing the parametric knowledge of LLMs and external documents to solve a multi-hop question. GenGround empowers LLMs to alternate two phases until the final answer is derived: (1) formulate a simpler, single-hop question and directly generate the answer; (2) ground the question-answer pair into retrieved documents, amending any wrong predictions in the answer. We also propose an instructional grounding distillation method to generalize our method into smaller models. Extensive experiments conducted on four datasets illustrate the superiority of our method. To further facilitate future research, we have collected a dataset that traces the reasoning process.
Anthology ID:
2024.acl-long.397
Volume:
Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
August
Year:
2024
Address:
Bangkok, Thailand
Editors:
Lun-Wei Ku, Andre Martins, Vivek Srikumar
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
7339–7353
Language:
URL:
https://aclanthology.org/2024.acl-long.397
DOI:
10.18653/v1/2024.acl-long.397
Bibkey:
Cite (ACL):
Zhengliang Shi, Shuo Zhang, Weiwei Sun, Shen Gao, Pengjie Ren, Zhumin Chen, and Zhaochun Ren. 2024. Generate-then-Ground in Retrieval-Augmented Generation for Multi-hop Question Answering. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 7339–7353, Bangkok, Thailand. Association for Computational Linguistics.
Cite (Informal):
Generate-then-Ground in Retrieval-Augmented Generation for Multi-hop Question Answering (Shi et al., ACL 2024)
Copy Citation:
PDF:
https://preview.aclanthology.org/landing_page/2024.acl-long.397.pdf