Saved in:
Bibliographic Details
Main Authors: Liu, Tianci, Jiang, Haoxiang, Wang, Tianze, Xu, Ran, Yu, Yue, Zhang, Linjun, Zhao, Tuo, Wang, Haoyu
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.10993
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915154167332864
author Liu, Tianci
Jiang, Haoxiang
Wang, Tianze
Xu, Ran
Yu, Yue
Zhang, Linjun
Zhao, Tuo
Wang, Haoyu
author_facet Liu, Tianci
Jiang, Haoxiang
Wang, Tianze
Xu, Ran
Yu, Yue
Zhang, Linjun
Zhao, Tuo
Wang, Haoyu
contents Large language models (LLMs) have achieved impressive performance but face high computational costs and latency, limiting their deployment in resource-constrained settings. In contrast, small-scale LLMs (SLMs) are more efficient yet struggle to capture evolving real-world knowledge. Retrieval-augmented generation (RAG) helps by integrating external knowledge, but imperfect retrieval can introduce distracting noise that misleads SLMs. We propose RoseRAG, a robust RAG framework for SLMs via Margin-aware Preference Optimization. RoseRAG employs multi-turn prompting for detailed reasoning, rejection sampling for high-quality explanations, and contrastive preference selection to refine responses by maximizing the likelihood gap between preferred and non-preferred outputs. By integrating these components into a margin-aware optimization process, RoseRAG robustly enhances the accuracy and reliability of SLMs for RAG applications. Extensive experiments on three open-domain question answering benchmarks indicate that our innovative RoseRAG surpasses state-of-the-art baselines significantly.
format Preprint
id arxiv_https___arxiv_org_abs_2502_10993
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle RoseRAG: Robust Retrieval-augmented Generation with Small-scale LLMs via Margin-aware Preference Optimization
Liu, Tianci
Jiang, Haoxiang
Wang, Tianze
Xu, Ran
Yu, Yue
Zhang, Linjun
Zhao, Tuo
Wang, Haoyu
Computation and Language
Machine Learning
Large language models (LLMs) have achieved impressive performance but face high computational costs and latency, limiting their deployment in resource-constrained settings. In contrast, small-scale LLMs (SLMs) are more efficient yet struggle to capture evolving real-world knowledge. Retrieval-augmented generation (RAG) helps by integrating external knowledge, but imperfect retrieval can introduce distracting noise that misleads SLMs. We propose RoseRAG, a robust RAG framework for SLMs via Margin-aware Preference Optimization. RoseRAG employs multi-turn prompting for detailed reasoning, rejection sampling for high-quality explanations, and contrastive preference selection to refine responses by maximizing the likelihood gap between preferred and non-preferred outputs. By integrating these components into a margin-aware optimization process, RoseRAG robustly enhances the accuracy and reliability of SLMs for RAG applications. Extensive experiments on three open-domain question answering benchmarks indicate that our innovative RoseRAG surpasses state-of-the-art baselines significantly.
title RoseRAG: Robust Retrieval-augmented Generation with Small-scale LLMs via Margin-aware Preference Optimization
topic Computation and Language
Machine Learning
url https://arxiv.org/abs/2502.10993