A MapReduce Approach to Effectively Utilize Long Context Information in Retrieval Augmented Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Gongbo, Xu, Zihan, Jin, Qiao, Chen, Fangyi, Fang, Yilu, Liu, Yi, Rousseau, Justin F., Xu, Ziyang, Lu, Zhiyong, Weng, Chunhua, Peng, Yifan
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913620455063552
author Zhang, Gongbo
Xu, Zihan
Jin, Qiao
Chen, Fangyi
Fang, Yilu
Liu, Yi
Rousseau, Justin F.
Xu, Ziyang
Lu, Zhiyong
Weng, Chunhua
Peng, Yifan
author_facet Zhang, Gongbo
Xu, Zihan
Jin, Qiao
Chen, Fangyi
Fang, Yilu
Liu, Yi
Rousseau, Justin F.
Xu, Ziyang
Lu, Zhiyong
Weng, Chunhua
Peng, Yifan
contents While holding great promise for improving and facilitating healthcare, large language models (LLMs) struggle to produce up-to-date responses on evolving topics due to outdated knowledge or hallucination. Retrieval-augmented generation (RAG) is a pivotal innovation that improves the accuracy and relevance of LLM responses by integrating LLMs with a search engine and external sources of knowledge. However, the quality of RAG responses can be largely impacted by the rank and density of key information in the retrieval results, such as the "lost-in-the-middle" problem. In this work, we aim to improve the robustness and reliability of the RAG workflow in the medical domain. Specifically, we propose a map-reduce strategy, BriefContext, to combat the "lost-in-the-middle" issue without modifying the model weights. We demonstrated the advantage of the workflow with various LLM backbones and on multiple QA datasets. This method promises to improve the safety and reliability of LLMs deployed in healthcare domains.
format Preprint
id arxiv_https___arxiv_org_abs_2412_15271
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A MapReduce Approach to Effectively Utilize Long Context Information in Retrieval Augmented Language Models
Zhang, Gongbo
Xu, Zihan
Jin, Qiao
Chen, Fangyi
Fang, Yilu
Liu, Yi
Rousseau, Justin F.
Xu, Ziyang
Lu, Zhiyong
Weng, Chunhua
Peng, Yifan
Computation and Language
Information Retrieval
While holding great promise for improving and facilitating healthcare, large language models (LLMs) struggle to produce up-to-date responses on evolving topics due to outdated knowledge or hallucination. Retrieval-augmented generation (RAG) is a pivotal innovation that improves the accuracy and relevance of LLM responses by integrating LLMs with a search engine and external sources of knowledge. However, the quality of RAG responses can be largely impacted by the rank and density of key information in the retrieval results, such as the "lost-in-the-middle" problem. In this work, we aim to improve the robustness and reliability of the RAG workflow in the medical domain. Specifically, we propose a map-reduce strategy, BriefContext, to combat the "lost-in-the-middle" issue without modifying the model weights. We demonstrated the advantage of the workflow with various LLM backbones and on multiple QA datasets. This method promises to improve the safety and reliability of LLMs deployed in healthcare domains.
title A MapReduce Approach to Effectively Utilize Long Context Information in Retrieval Augmented Language Models
topic Computation and Language
Information Retrieval
url https://arxiv.org/abs/2412.15271