Logo des Repositoriums
 

ATDLLMD: Acceptance test-driven LLM development

dc.contributor.authorFaragó, David
dc.contributor.editorHerrmann, Andrea
dc.date.accessioned2024-07-26T10:37:42Z
dc.date.available2024-07-26T10:37:42Z
dc.date.issued2024
dc.description.abstractSince the capabilities of Large Language Models (LLMs) have massively increased in the last years, many new applications based on LLMs are possible. However, these new applications also pose new challenges in LLM development. This article proposes an acceptance test-driven development (ATDD) style, baptized ATDLLMD, where the LLM’s training and test sets are extended in each iteration by data coming from validation of the previous iteration’s LLM and system around the LLM. So the validation phase supplies the additional or updated data for training and verification of the LLM. ATDLLMD is made possible by two major innovative solutions: applying the innovative CPMAI process, and applying our own verification tool, LM-Eval, leading to a red-train green cycle for LLM development, which resembles ATDD, but integrates data science best practices.en
dc.identifier.urihttps://dl.gi.de/handle/20.500.12116/44202
dc.language.isoen
dc.pubPlaceBonn
dc.publisherGesellschaft für Informatik e.V.
dc.relation.ispartofSoftwaretechnik-Trends Band 44, Heft 2
dc.relation.ispartofseriesSoftwaretechnik-Trends
dc.subjectLarge Language Model
dc.subjectLLM
dc.subjectdevelopment process
dc.subjecttest-first
dc.subjectLLM evaluation
dc.subjectLLM testing
dc.subjectdata-centric AI
dc.subjectbusiness-centric AI
dc.titleATDLLMD: Acceptance test-driven LLM developmenten
dc.typeText/Conference Paper
mci.conference.date15.-16. Februar 2024
mci.conference.locationGummersbach
mci.conference.sessiontitleTreffen der GI-Fachgruppe Test, Analyse und Verifikation von Software (TAV 49)
mci.reference.pages13-17

Dateien

Originalbündel
1 - 1 von 1
Vorschaubild nicht verfügbar
Name:
2_TAV49_Farago.pdf
Größe:
1 MB
Format:
Adobe Portable Document Format