@inproceedings{weller-etal-2024-according,
title = "{``}According to . . . {''}: Prompting Language Models Improves Quoting from Pre-Training Data",
author = "Weller, Orion and
Marone, Marc and
Weir, Nathaniel and
Lawrie, Dawn and
Khashabi, Daniel and
Van Durme, Benjamin",
editor = "Graham, Yvette and
Purver, Matthew",
booktitle = "Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = mar,
year = "2024",
address = "St. Julian{'}s, Malta",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.eacl-long.140",
pages = "2288--2301",
abstract = "Large Language Models (LLMs) may hallucinate and generate fake information, despite pre-training on factual data. Inspired by the journalistic device of {``}according to sources{''}, we propose according-to prompting: directing LLMs to ground responses against previously observed text. To quantify this grounding, we propose a novel evaluation metric (QUIP-Score) that measures the extent to which model-produced answers are directly found in underlying text corpora. We illustrate with experiments on three corpora (Wikipedia, PubMed, and the U.S. legal tax code) that these prompts improve grounding under our metrics, with the additional benefit of often improving end-task performance. Furthermore, prompts that ask the model to decrease grounding (or to ground to other corpora) indeed decrease QUIP-Score, indicating the ability of LLMs to increase or decrease grounded generations on request.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="weller-etal-2024-according">
<titleInfo>
<title>“According to . . . ”: Prompting Language Models Improves Quoting from Pre-Training Data</title>
</titleInfo>
<name type="personal">
<namePart type="given">Orion</namePart>
<namePart type="family">Weller</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marc</namePart>
<namePart type="family">Marone</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nathaniel</namePart>
<namePart type="family">Weir</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dawn</namePart>
<namePart type="family">Lawrie</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Khashabi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Benjamin</namePart>
<namePart type="family">Van Durme</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-03</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yvette</namePart>
<namePart type="family">Graham</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Matthew</namePart>
<namePart type="family">Purver</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">St. Julian’s, Malta</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Large Language Models (LLMs) may hallucinate and generate fake information, despite pre-training on factual data. Inspired by the journalistic device of “according to sources”, we propose according-to prompting: directing LLMs to ground responses against previously observed text. To quantify this grounding, we propose a novel evaluation metric (QUIP-Score) that measures the extent to which model-produced answers are directly found in underlying text corpora. We illustrate with experiments on three corpora (Wikipedia, PubMed, and the U.S. legal tax code) that these prompts improve grounding under our metrics, with the additional benefit of often improving end-task performance. Furthermore, prompts that ask the model to decrease grounding (or to ground to other corpora) indeed decrease QUIP-Score, indicating the ability of LLMs to increase or decrease grounded generations on request.</abstract>
<identifier type="citekey">weller-etal-2024-according</identifier>
<location>
<url>https://aclanthology.org/2024.eacl-long.140</url>
</location>
<part>
<date>2024-03</date>
<extent unit="page">
<start>2288</start>
<end>2301</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T “According to . . . ”: Prompting Language Models Improves Quoting from Pre-Training Data
%A Weller, Orion
%A Marone, Marc
%A Weir, Nathaniel
%A Lawrie, Dawn
%A Khashabi, Daniel
%A Van Durme, Benjamin
%Y Graham, Yvette
%Y Purver, Matthew
%S Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2024
%8 March
%I Association for Computational Linguistics
%C St. Julian’s, Malta
%F weller-etal-2024-according
%X Large Language Models (LLMs) may hallucinate and generate fake information, despite pre-training on factual data. Inspired by the journalistic device of “according to sources”, we propose according-to prompting: directing LLMs to ground responses against previously observed text. To quantify this grounding, we propose a novel evaluation metric (QUIP-Score) that measures the extent to which model-produced answers are directly found in underlying text corpora. We illustrate with experiments on three corpora (Wikipedia, PubMed, and the U.S. legal tax code) that these prompts improve grounding under our metrics, with the additional benefit of often improving end-task performance. Furthermore, prompts that ask the model to decrease grounding (or to ground to other corpora) indeed decrease QUIP-Score, indicating the ability of LLMs to increase or decrease grounded generations on request.
%U https://aclanthology.org/2024.eacl-long.140
%P 2288-2301
Markdown (Informal)
[“According to . . . ”: Prompting Language Models Improves Quoting from Pre-Training Data](https://aclanthology.org/2024.eacl-long.140) (Weller et al., EACL 2024)
ACL