UncertaintyRAG: Span-Level Uncertainty Enhanced Long-Context Modeling for Retrieval-Augmented Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Zixuan, Xiong, Jing, Ye, Fanghua, Zheng, Chuanyang, Wu, Xun, Lu, Jianqiao, Wan, Zhongwei, Liang, Xiaodan, Li, Chengming, Sun, Zhenan, Kong, Lingpeng, Wong, Ngai
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910631329792000
author Li, Zixuan
Xiong, Jing
Ye, Fanghua
Zheng, Chuanyang
Wu, Xun
Lu, Jianqiao
Wan, Zhongwei
Liang, Xiaodan
Li, Chengming
Sun, Zhenan
Kong, Lingpeng
Wong, Ngai
author_facet Li, Zixuan
Xiong, Jing
Ye, Fanghua
Zheng, Chuanyang
Wu, Xun
Lu, Jianqiao
Wan, Zhongwei
Liang, Xiaodan
Li, Chengming
Sun, Zhenan
Kong, Lingpeng
Wong, Ngai
contents We present UncertaintyRAG, a novel approach for long-context Retrieval-Augmented Generation (RAG) that utilizes Signal-to-Noise Ratio (SNR)-based span uncertainty to estimate similarity between text chunks. This span uncertainty enhances model calibration, improving robustness and mitigating semantic inconsistencies introduced by random chunking. Leveraging this insight, we propose an efficient unsupervised learning technique to train the retrieval model, alongside an effective data sampling and scaling strategy. UncertaintyRAG outperforms baselines by 2.03% on LLaMA-2-7B, achieving state-of-the-art results while using only 4% of the training data compared to other advanced open-source retrieval models under distribution shift settings. Our method demonstrates strong calibration through span uncertainty, leading to improved generalization and robustness in long-context RAG tasks. Additionally, UncertaintyRAG provides a lightweight retrieval model that can be integrated into any large language model with varying context window lengths, without the need for fine-tuning, showcasing the flexibility of our approach.
format Preprint
id arxiv_https___arxiv_org_abs_2410_02719
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle UncertaintyRAG: Span-Level Uncertainty Enhanced Long-Context Modeling for Retrieval-Augmented Generation
Li, Zixuan
Xiong, Jing
Ye, Fanghua
Zheng, Chuanyang
Wu, Xun
Lu, Jianqiao
Wan, Zhongwei
Liang, Xiaodan
Li, Chengming
Sun, Zhenan
Kong, Lingpeng
Wong, Ngai
Computation and Language
We present UncertaintyRAG, a novel approach for long-context Retrieval-Augmented Generation (RAG) that utilizes Signal-to-Noise Ratio (SNR)-based span uncertainty to estimate similarity between text chunks. This span uncertainty enhances model calibration, improving robustness and mitigating semantic inconsistencies introduced by random chunking. Leveraging this insight, we propose an efficient unsupervised learning technique to train the retrieval model, alongside an effective data sampling and scaling strategy. UncertaintyRAG outperforms baselines by 2.03% on LLaMA-2-7B, achieving state-of-the-art results while using only 4% of the training data compared to other advanced open-source retrieval models under distribution shift settings. Our method demonstrates strong calibration through span uncertainty, leading to improved generalization and robustness in long-context RAG tasks. Additionally, UncertaintyRAG provides a lightweight retrieval model that can be integrated into any large language model with varying context window lengths, without the need for fine-tuning, showcasing the flexibility of our approach.
title UncertaintyRAG: Span-Level Uncertainty Enhanced Long-Context Modeling for Retrieval-Augmented Generation
topic Computation and Language
url https://arxiv.org/abs/2410.02719