Within-Model vs Between-Prompt Variability in Large Language Models for Creative Tasks

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Haase, Jennifer, Gonnermann-Müller, Jana, Hanel, Paul H. P., Leins, Nicolas, Kosch, Thomas, Mendling, Jan, Pokutta, Sebastian
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918312959541248
author Haase, Jennifer
Gonnermann-Müller, Jana
Hanel, Paul H. P.
Leins, Nicolas
Kosch, Thomas
Mendling, Jan
Pokutta, Sebastian
author_facet Haase, Jennifer
Gonnermann-Müller, Jana
Hanel, Paul H. P.
Leins, Nicolas
Kosch, Thomas
Mendling, Jan
Pokutta, Sebastian
contents How much of LLM output variance is explained by prompts versus model choice versus stochasticity through sampling? We answer this by evaluating 12 LLMs on 10 creativity prompts with 100 samples each (N = 12,000). For output quality (originality), prompts explain 36.43% of variance, comparable to model choice (40.94%). But for output quantity (fluency), model choice (51.25%) and within-LLM variance (33.70%) dominate, with prompts explaining only 4.22%. Prompts are powerful levers for steering output quality, but given the substantial within-LLM variance (10-34%), single-sample evaluations risk conflating sampling noise with genuine prompt or model effects.
format Preprint
id arxiv_https___arxiv_org_abs_2601_21339
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Within-Model vs Between-Prompt Variability in Large Language Models for Creative Tasks
Haase, Jennifer
Gonnermann-Müller, Jana
Hanel, Paul H. P.
Leins, Nicolas
Kosch, Thomas
Mendling, Jan
Pokutta, Sebastian
Artificial Intelligence
How much of LLM output variance is explained by prompts versus model choice versus stochasticity through sampling? We answer this by evaluating 12 LLMs on 10 creativity prompts with 100 samples each (N = 12,000). For output quality (originality), prompts explain 36.43% of variance, comparable to model choice (40.94%). But for output quantity (fluency), model choice (51.25%) and within-LLM variance (33.70%) dominate, with prompts explaining only 4.22%. Prompts are powerful levers for steering output quality, but given the substantial within-LLM variance (10-34%), single-sample evaluations risk conflating sampling noise with genuine prompt or model effects.
title Within-Model vs Between-Prompt Variability in Large Language Models for Creative Tasks
topic Artificial Intelligence
url https://arxiv.org/abs/2601.21339