@inproceedings{mishra-etal-2024-characterizing,
title = "Characterizing Large Language Models as Rationalizers of Knowledge-intensive Tasks",
author = "Mishra, Aditi and
Rahman, Sajjadur and
Mitra, Kushan and
Kim, Hannah and
Hruschka, Estevam",
editor = "Ku, Lun-Wei and
Martins, Andre and
Srikumar, Vivek",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2024",
month = aug,
year = "2024",
address = "Bangkok, Thailand",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-acl.484/",
doi = "10.18653/v1/2024.findings-acl.484",
pages = "8117--8139",
abstract = "Large language models (LLMs) are proficient at generating fluent text with minimal task-specific supervision. However, their ability to generate rationales for knowledge-intensive tasks (KITs) remains under-explored. Generating rationales for KIT solutions, such as commonsense multiple-choice QA, requires external knowledge to support predictions and refute alternate options. In this work, we consider the task of generating retrieval-augmented rationalization of KIT model predictions via external knowledge guidance within a few-shot setting. Surprisingly, crowd-workers preferred LLM-generated rationales over existing crowd-sourced rationales, generated in a similar knowledge-guided setting, on aspects such as factuality, sufficiency, and convincingness. However, fine-grained evaluation of such rationales highlights the need for further improvements in conciseness, novelty, and domain invariance. Additionally, through an expert-sourced study evaluating the reliability of the rationales, we demonstrate that humans' trust in LLM-generated rationales erodes when communicated faithfully, i.e., without taking model prediction accuracy into account. We find that even instrumenting simple guardrails can be effective for reliable rationalization."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="mishra-etal-2024-characterizing">
<titleInfo>
<title>Characterizing Large Language Models as Rationalizers of Knowledge-intensive Tasks</title>
</titleInfo>
<name type="personal">
<namePart type="given">Aditi</namePart>
<namePart type="family">Mishra</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sajjadur</namePart>
<namePart type="family">Rahman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kushan</namePart>
<namePart type="family">Mitra</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hannah</namePart>
<namePart type="family">Kim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Estevam</namePart>
<namePart type="family">Hruschka</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2024</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lun-Wei</namePart>
<namePart type="family">Ku</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andre</namePart>
<namePart type="family">Martins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivek</namePart>
<namePart type="family">Srikumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bangkok, Thailand</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Large language models (LLMs) are proficient at generating fluent text with minimal task-specific supervision. However, their ability to generate rationales for knowledge-intensive tasks (KITs) remains under-explored. Generating rationales for KIT solutions, such as commonsense multiple-choice QA, requires external knowledge to support predictions and refute alternate options. In this work, we consider the task of generating retrieval-augmented rationalization of KIT model predictions via external knowledge guidance within a few-shot setting. Surprisingly, crowd-workers preferred LLM-generated rationales over existing crowd-sourced rationales, generated in a similar knowledge-guided setting, on aspects such as factuality, sufficiency, and convincingness. However, fine-grained evaluation of such rationales highlights the need for further improvements in conciseness, novelty, and domain invariance. Additionally, through an expert-sourced study evaluating the reliability of the rationales, we demonstrate that humans’ trust in LLM-generated rationales erodes when communicated faithfully, i.e., without taking model prediction accuracy into account. We find that even instrumenting simple guardrails can be effective for reliable rationalization.</abstract>
<identifier type="citekey">mishra-etal-2024-characterizing</identifier>
<identifier type="doi">10.18653/v1/2024.findings-acl.484</identifier>
<location>
<url>https://aclanthology.org/2024.findings-acl.484/</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>8117</start>
<end>8139</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Characterizing Large Language Models as Rationalizers of Knowledge-intensive Tasks
%A Mishra, Aditi
%A Rahman, Sajjadur
%A Mitra, Kushan
%A Kim, Hannah
%A Hruschka, Estevam
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Findings of the Association for Computational Linguistics: ACL 2024
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand
%F mishra-etal-2024-characterizing
%X Large language models (LLMs) are proficient at generating fluent text with minimal task-specific supervision. However, their ability to generate rationales for knowledge-intensive tasks (KITs) remains under-explored. Generating rationales for KIT solutions, such as commonsense multiple-choice QA, requires external knowledge to support predictions and refute alternate options. In this work, we consider the task of generating retrieval-augmented rationalization of KIT model predictions via external knowledge guidance within a few-shot setting. Surprisingly, crowd-workers preferred LLM-generated rationales over existing crowd-sourced rationales, generated in a similar knowledge-guided setting, on aspects such as factuality, sufficiency, and convincingness. However, fine-grained evaluation of such rationales highlights the need for further improvements in conciseness, novelty, and domain invariance. Additionally, through an expert-sourced study evaluating the reliability of the rationales, we demonstrate that humans’ trust in LLM-generated rationales erodes when communicated faithfully, i.e., without taking model prediction accuracy into account. We find that even instrumenting simple guardrails can be effective for reliable rationalization.
%R 10.18653/v1/2024.findings-acl.484
%U https://aclanthology.org/2024.findings-acl.484/
%U https://doi.org/10.18653/v1/2024.findings-acl.484
%P 8117-8139
Markdown (Informal)
[Characterizing Large Language Models as Rationalizers of Knowledge-intensive Tasks](https://aclanthology.org/2024.findings-acl.484/) (Mishra et al., Findings 2024)
ACL