Saved in:
Bibliographic Details
Main Authors: Dai, Lu, Sun, Liang, Cao, Fanpu, Rao, Ziyang, Yang, Cehao, Liu, Hao, Xiong, Hui
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.00505
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918507588878336
author Dai, Lu
Sun, Liang
Cao, Fanpu
Rao, Ziyang
Yang, Cehao
Liu, Hao
Xiong, Hui
author_facet Dai, Lu
Sun, Liang
Cao, Fanpu
Rao, Ziyang
Yang, Cehao
Liu, Hao
Xiong, Hui
contents Modern information retrieval (IR) is no longer consumed primarily by humans but increasingly by large language models (LLMs) via retrieval-augmented generation (RAG) and agentic search. Unlike human users, LLMs are constrained by limited attention budgets and are uniquely vulnerable to noise; misleading or irrelevant information is no longer just a nuisance, but a direct cause of hallucinations and reasoning failures. In this perspective paper, we argue that denoising-maximizing usable evidence density and verifiability within a context window-is becoming the primary bottleneck across the full information access pipeline. We conceptualize this paradigm shift through a four-stage framework of IR challenges: from inaccessible to undiscoverable, to misaligned, and finally to unverifiable. Furthermore, we provide a pipeline-organized taxonomy of signal-to-noise optimization techniques, spanning indexing, retrieval, context engineering, verification, and agentic workflow. We also present research works on information denoising in domains that rely heavily on retrieval such as lifelong assistant, coding agent, deep research, and multimodal understanding.
format Preprint
id arxiv_https___arxiv_org_abs_2605_00505
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle LLM-Oriented Information Retrieval: A Denoising-First Perspective
Dai, Lu
Sun, Liang
Cao, Fanpu
Rao, Ziyang
Yang, Cehao
Liu, Hao
Xiong, Hui
Information Retrieval
Artificial Intelligence
Computation and Language
Modern information retrieval (IR) is no longer consumed primarily by humans but increasingly by large language models (LLMs) via retrieval-augmented generation (RAG) and agentic search. Unlike human users, LLMs are constrained by limited attention budgets and are uniquely vulnerable to noise; misleading or irrelevant information is no longer just a nuisance, but a direct cause of hallucinations and reasoning failures. In this perspective paper, we argue that denoising-maximizing usable evidence density and verifiability within a context window-is becoming the primary bottleneck across the full information access pipeline. We conceptualize this paradigm shift through a four-stage framework of IR challenges: from inaccessible to undiscoverable, to misaligned, and finally to unverifiable. Furthermore, we provide a pipeline-organized taxonomy of signal-to-noise optimization techniques, spanning indexing, retrieval, context engineering, verification, and agentic workflow. We also present research works on information denoising in domains that rely heavily on retrieval such as lifelong assistant, coding agent, deep research, and multimodal understanding.
title LLM-Oriented Information Retrieval: A Denoising-First Perspective
topic Information Retrieval
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2605.00505