ARL2: Aligning Retrievers with Black-box Large Language Models via Self-guided Adaptive Relevance Labeling

LingXi Zhang, Yue Yu, Kuan Wang, Chao Zhang

Abstract
Retrieval-augmented generation enhances large language models (LLMs) by incorporating relevant information from external knowledge sources. This enables LLMs to adapt to specific domains and mitigate hallucinations in knowledge-intensive tasks. However, existing retrievers are often misaligned with LLMs due to separate training processes and the inherent black-box nature of LLMs. To address this challenge, we propose ARL2, a retriever learning technique that harnesses LLMs as labelers. ARL2 leverages LLMs to annotate and score adaptive relevance evidence, enabling the retriever to learn from robust LLM supervision. Furthermore, ARL2 incorporates a self-training strategy to minimize the cost of API calls. Extensive experiments demonstrate the effectiveness of ARL2, achieving accuracy improvements of 5.4% on NQ and 4.6% on MMLU compared to the state-of-the-art methods. Additionally, ARL2 exhibits robust transfer learning capabilities and strong zero-shot generalization abilities.
Anthology ID:
2024.acl-long.203
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:
3708–3719
Language:
URL:
https://aclanthology.org/2024.acl-long.203
DOI:
Bibkey:
Cite (ACL):
LingXi Zhang, Yue Yu, Kuan Wang, and Chao Zhang. 2024. ARL2: Aligning Retrievers with Black-box Large Language Models via Self-guided Adaptive Relevance Labeling. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3708–3719, Bangkok, Thailand. Association for Computational Linguistics.
Cite (Informal):
ARL2: Aligning Retrievers with Black-box Large Language Models via Self-guided Adaptive Relevance Labeling (Zhang et al., ACL 2024)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-4/2024.acl-long.203.pdf
PDF Search