@inproceedings{cuconasu-etal-2026-tale,
title = "A Tale of Trust and Accuracy: Base vs. Instruct {LLM}s in {RAG} Systems",
author = "Cuconasu, Florin and
Trappolini, Giovanni and
Tonellotto, Nicola and
Silvestri, Fabrizio",
editor = "Yang, Eugene and
Lawrie, Dawn and
MacAvaney, Sean and
Mayfield, James and
Soldaini, Luca and
Yates, Andrew",
booktitle = "Proceedings of the 1st Workshop on Multilingual Report Generation via Retrieval Augmented Generation ({RAG}4{R}eports 2026)",
month = jul,
year = "2026",
address = "San Diego, CA, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.rag4reports-1.3/",
pages = "1--23",
ISBN = "979-8-89176-417-0",
abstract = "Retrieval-Augmented Generation (RAG) represents a significant advancement in artificial intelligence combining a retrieval phase with a generative phase, with the latter typically being powered by Large Language Models (LLMs). Common wisdom and practices in RAG involve using ``instructed'' LLMs, which are fine-tuned with supervised training to enhance their ability to follow instructions and are aligned with human preferences using state-of-the-art techniques.However, contrary to this popular belief, our study demonstrates that base models outperform their instructed counterparts in RAG tasks by 20{\%} on average under our experimental settings. This finding challenges the prevailing assumptions about the superiority of instructed LLMs in RAG applications. Further investigations reveal a more complex situation, questioning fundamental aspects of RAG and suggesting the need for broader discussions on the topic; or, as Fromm would have it, ``Seldom is a glance at the statistics enough to understand the meaning of the figures''."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="cuconasu-etal-2026-tale">
<titleInfo>
<title>A Tale of Trust and Accuracy: Base vs. Instruct LLMs in RAG Systems</title>
</titleInfo>
<name type="personal">
<namePart type="given">Florin</namePart>
<namePart type="family">Cuconasu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Giovanni</namePart>
<namePart type="family">Trappolini</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nicola</namePart>
<namePart type="family">Tonellotto</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fabrizio</namePart>
<namePart type="family">Silvestri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 1st Workshop on Multilingual Report Generation via Retrieval Augmented Generation (RAG4Reports 2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Eugene</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dawn</namePart>
<namePart type="family">Lawrie</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sean</namePart>
<namePart type="family">MacAvaney</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">James</namePart>
<namePart type="family">Mayfield</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Luca</namePart>
<namePart type="family">Soldaini</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andrew</namePart>
<namePart type="family">Yates</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, CA, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-417-0</identifier>
</relatedItem>
<abstract>Retrieval-Augmented Generation (RAG) represents a significant advancement in artificial intelligence combining a retrieval phase with a generative phase, with the latter typically being powered by Large Language Models (LLMs). Common wisdom and practices in RAG involve using “instructed” LLMs, which are fine-tuned with supervised training to enhance their ability to follow instructions and are aligned with human preferences using state-of-the-art techniques.However, contrary to this popular belief, our study demonstrates that base models outperform their instructed counterparts in RAG tasks by 20% on average under our experimental settings. This finding challenges the prevailing assumptions about the superiority of instructed LLMs in RAG applications. Further investigations reveal a more complex situation, questioning fundamental aspects of RAG and suggesting the need for broader discussions on the topic; or, as Fromm would have it, “Seldom is a glance at the statistics enough to understand the meaning of the figures”.</abstract>
<identifier type="citekey">cuconasu-etal-2026-tale</identifier>
<location>
<url>https://aclanthology.org/2026.rag4reports-1.3/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>1</start>
<end>23</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T A Tale of Trust and Accuracy: Base vs. Instruct LLMs in RAG Systems
%A Cuconasu, Florin
%A Trappolini, Giovanni
%A Tonellotto, Nicola
%A Silvestri, Fabrizio
%Y Yang, Eugene
%Y Lawrie, Dawn
%Y MacAvaney, Sean
%Y Mayfield, James
%Y Soldaini, Luca
%Y Yates, Andrew
%S Proceedings of the 1st Workshop on Multilingual Report Generation via Retrieval Augmented Generation (RAG4Reports 2026)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, CA, USA
%@ 979-8-89176-417-0
%F cuconasu-etal-2026-tale
%X Retrieval-Augmented Generation (RAG) represents a significant advancement in artificial intelligence combining a retrieval phase with a generative phase, with the latter typically being powered by Large Language Models (LLMs). Common wisdom and practices in RAG involve using “instructed” LLMs, which are fine-tuned with supervised training to enhance their ability to follow instructions and are aligned with human preferences using state-of-the-art techniques.However, contrary to this popular belief, our study demonstrates that base models outperform their instructed counterparts in RAG tasks by 20% on average under our experimental settings. This finding challenges the prevailing assumptions about the superiority of instructed LLMs in RAG applications. Further investigations reveal a more complex situation, questioning fundamental aspects of RAG and suggesting the need for broader discussions on the topic; or, as Fromm would have it, “Seldom is a glance at the statistics enough to understand the meaning of the figures”.
%U https://aclanthology.org/2026.rag4reports-1.3/
%P 1-23
Markdown (Informal)
[A Tale of Trust and Accuracy: Base vs. Instruct LLMs in RAG Systems](https://aclanthology.org/2026.rag4reports-1.3/) (Cuconasu et al., RAG4Reports 2026)
ACL
- Florin Cuconasu, Giovanni Trappolini, Nicola Tonellotto, and Fabrizio Silvestri. 2026. A Tale of Trust and Accuracy: Base vs. Instruct LLMs in RAG Systems. In Proceedings of the 1st Workshop on Multilingual Report Generation via Retrieval Augmented Generation (RAG4Reports 2026), pages 1–23, San Diego, CA, USA. Association for Computational Linguistics.