Unsupervised Query Routing for Retrieval Augmented Generation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916565824307200 |
|---|---|
| author | Mu, Feiteng Zhang, Liwen Jiang, Yong Li, Wenjie Zhang, Zhen Xie, Pengjun Huang, Fei |
| author_facet | Mu, Feiteng Zhang, Liwen Jiang, Yong Li, Wenjie Zhang, Zhen Xie, Pengjun Huang, Fei |
| contents | Query routing for retrieval-augmented generation aims to assign an input query to the most suitable search engine. Existing works rely heavily on supervised datasets that require extensive manual annotation, resulting in high costs and limited scalability, as well as poor generalization to out-of-distribution scenarios. To address these challenges, we introduce a novel unsupervised method that constructs the "upper-bound" response to evaluate the quality of retrieval-augmented responses. This evaluation enables the decision of the most suitable search engine for a given query. By eliminating manual annotations, our approach can automatically process large-scale real user queries and create training data. We conduct extensive experiments across five datasets, demonstrating that our method significantly enhances scalability and generalization capabilities. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_07793 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Unsupervised Query Routing for Retrieval Augmented Generation Mu, Feiteng Zhang, Liwen Jiang, Yong Li, Wenjie Zhang, Zhen Xie, Pengjun Huang, Fei Information Retrieval Query routing for retrieval-augmented generation aims to assign an input query to the most suitable search engine. Existing works rely heavily on supervised datasets that require extensive manual annotation, resulting in high costs and limited scalability, as well as poor generalization to out-of-distribution scenarios. To address these challenges, we introduce a novel unsupervised method that constructs the "upper-bound" response to evaluate the quality of retrieval-augmented responses. This evaluation enables the decision of the most suitable search engine for a given query. By eliminating manual annotations, our approach can automatically process large-scale real user queries and create training data. We conduct extensive experiments across five datasets, demonstrating that our method significantly enhances scalability and generalization capabilities. |
| title | Unsupervised Query Routing for Retrieval Augmented Generation |
| topic | Information Retrieval |
| url | https://arxiv.org/abs/2501.07793 |