ARL2: Aligning Retrievers for Black-box Large Language Models via Self-guided Adaptive Relevance Labeling
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911901444734976 |
|---|---|
| author | Zhang, Lingxi Yu, Yue Wang, Kuan Zhang, Chao |
| author_facet | Zhang, Lingxi Yu, Yue Wang, Kuan Zhang, Chao |
| contents | 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 their separate training processes and the 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 relevant evidence, enabling learning the retriever from robust LLM supervision. Furthermore, ARL2 uses an adaptive self-training strategy for curating high-quality and diverse relevance data, which can effectively reduce the annotation cost. 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. Our code will be published at \url{https://github.com/zhanglingxi-cs/ARL2}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_13542 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | ARL2: Aligning Retrievers for Black-box Large Language Models via Self-guided Adaptive Relevance Labeling Zhang, Lingxi Yu, Yue Wang, Kuan Zhang, Chao Computation and Language Artificial Intelligence Information Retrieval Machine Learning 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 their separate training processes and the 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 relevant evidence, enabling learning the retriever from robust LLM supervision. Furthermore, ARL2 uses an adaptive self-training strategy for curating high-quality and diverse relevance data, which can effectively reduce the annotation cost. 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. Our code will be published at \url{https://github.com/zhanglingxi-cs/ARL2}. |
| title | ARL2: Aligning Retrievers for Black-box Large Language Models via Self-guided Adaptive Relevance Labeling |
| topic | Computation and Language Artificial Intelligence Information Retrieval Machine Learning |
| url | https://arxiv.org/abs/2402.13542 |