Designing Staged Evaluation Workflows for LLMs: Integrating Domain Experts, Lay Users, and Model-Generated Evaluation Criteria

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Szymanski, Annalisa, Gebreegziabher, Simret Araya, Anuyah, Oghenemaro, Metoyer, Ronald A., Li, Toby Jia-Jun
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917274923827200
author Szymanski, Annalisa
Gebreegziabher, Simret Araya
Anuyah, Oghenemaro
Metoyer, Ronald A.
Li, Toby Jia-Jun
author_facet Szymanski, Annalisa
Gebreegziabher, Simret Araya
Anuyah, Oghenemaro
Metoyer, Ronald A.
Li, Toby Jia-Jun
contents Large Language Models (LLMs) are increasingly utilized for domain-specific tasks, yet evaluating their outputs remains challenging. A common strategy is to apply evaluation criteria to assess alignment with domain-specific standards, yet little is understood about how criteria differ across sources or where each type is most useful in the evaluation process. This study investigates criteria developed by domain experts, lay users, and LLMs to identify their complementary roles within an evaluation workflow. Results show that experts produce fact-based criteria with long-term value, lay users emphasize usability with a shorter-term focus, and LLMs target procedural checks for immediate task requirements. We also examine how criteria evolve between a priori and a posteriori phases, noting drift across stages as well as convergence in the a posteriori phase. Based on our observations, we propose design guidelines for a staged evaluation workflow combining the complementary strengths of these sources to balance quality, cost, and scalability.
format Preprint
id arxiv_https___arxiv_org_abs_2410_02054
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Designing Staged Evaluation Workflows for LLMs: Integrating Domain Experts, Lay Users, and Model-Generated Evaluation Criteria
Szymanski, Annalisa
Gebreegziabher, Simret Araya
Anuyah, Oghenemaro
Metoyer, Ronald A.
Li, Toby Jia-Jun
Human-Computer Interaction
Large Language Models (LLMs) are increasingly utilized for domain-specific tasks, yet evaluating their outputs remains challenging. A common strategy is to apply evaluation criteria to assess alignment with domain-specific standards, yet little is understood about how criteria differ across sources or where each type is most useful in the evaluation process. This study investigates criteria developed by domain experts, lay users, and LLMs to identify their complementary roles within an evaluation workflow. Results show that experts produce fact-based criteria with long-term value, lay users emphasize usability with a shorter-term focus, and LLMs target procedural checks for immediate task requirements. We also examine how criteria evolve between a priori and a posteriori phases, noting drift across stages as well as convergence in the a posteriori phase. Based on our observations, we propose design guidelines for a staged evaluation workflow combining the complementary strengths of these sources to balance quality, cost, and scalability.
title Designing Staged Evaluation Workflows for LLMs: Integrating Domain Experts, Lay Users, and Model-Generated Evaluation Criteria
topic Human-Computer Interaction
url https://arxiv.org/abs/2410.02054