<div class="csl-bib-body">
<div class="csl-entry">Recski, G. (2024, November 19). <i>Fact-checking LLMs with explainable information extraction</i> [Conference Presentation]. Language Intelligence 2024, Austria. https://doi.org/10.34726/8540</div>
</div>
-
dc.identifier.uri
http://hdl.handle.net/20.500.12708/211064
-
dc.identifier.uri
https://doi.org/10.34726/8540
-
dc.description.abstract
Large Language Models (LLMs) have become the most commonly used tool in natural language processing (NLP), but practitioners have quickly discovered their shortcomings. LLMs make mistakes that we can neither predict nor prevent. They are black boxes whose behavior cannot be configured or explained. Their exact training process is unknown and they are prone to unwanted bias. Their computational cost is enormous, creating both financial and environmental burdens. And unless they are developed in house, their use raises privacy and security issues. All these problems severely limit the applicability of LLMs in domains that require high degrees of transparency and trustworthiness, such as legal, medical, or financial NLP applications. The NLP group at TU Wien has developed POTATO, the explainable information extraction framework. Validated across complex domains and multiple languages, POTATO is an open-source engine for rule-based information extraction that can be used either as an alternative or an addition to LLM-based solutions. In particular, POTATO can be deployed as a fact-checker for Retrieval Augmented Generation (RAG) systems that answer user queries based on reliable and relevant documents.
en
dc.language.iso
en
-
dc.rights.uri
http://creativecommons.org/licenses/by-sa/4.0/
-
dc.subject
LLM
en
dc.subject
Explainability
en
dc.subject
Information Extraction
en
dc.title
Fact-checking LLMs with explainable information extraction
en
dc.type
Presentation
en
dc.type
Vortrag
de
dc.rights.license
Creative Commons Namensnennung - Weitergabe unter gleichen Bedingungen 4.0 International
de
dc.rights.license
Creative Commons Attribution-ShareAlike 4.0 International
en
dc.identifier.doi
10.34726/8540
-
dc.type.category
Conference Presentation
-
tuw.publication.invited
invited
-
tuw.researchTopic.id
I4
-
tuw.researchTopic.id
X1
-
tuw.researchTopic.name
Information Systems Engineering
-
tuw.researchTopic.name
Beyond TUW-research focus
-
tuw.researchTopic.value
70
-
tuw.researchTopic.value
30
-
tuw.publication.orgunit
E194-04 - Forschungsbereich Data Science
-
tuw.author.orcid
0000-0001-5551-3100
-
dc.rights.identifier
CC BY-SA 4.0
de
dc.rights.identifier
CC BY-SA 4.0
en
tuw.event.name
Language Intelligence 2024
en
tuw.event.startdate
19-11-2024
-
tuw.event.enddate
20-11-2024
-
tuw.event.online
On Site
-
tuw.event.type
Event for non-scientific audience
-
tuw.event.country
AT
-
tuw.event.presenter
Recski, Gábor
-
tuw.event.track
Single Track
-
wb.sciencebranch
Sprach- und Literaturwissenschaften
-
wb.sciencebranch
Informatik
-
wb.sciencebranch.oefos
6020
-
wb.sciencebranch.oefos
1020
-
wb.sciencebranch.value
30
-
wb.sciencebranch.value
70
-
item.grantfulltext
open
-
item.languageiso639-1
en
-
item.fulltext
with Fulltext
-
item.openaccessfulltext
Open Access
-
item.cerifentitytype
Publications
-
item.openairecristype
http://purl.org/coar/resource_type/c_18cp
-
item.openairetype
conference paper not in proceedings
-
item.mimetype
application/pdf
-
crisitem.author.dept
E194-04 - Forschungsbereich Data Science
-
crisitem.author.orcid
0000-0001-5551-3100
-
crisitem.author.parentorg
E194 - Institut für Information Systems Engineering
