Consistency over accuracy: run-to-run stability of contemporary large language models on Turkish curriculum-aligned theoretical anatomy multiple-choice questions
Yazarlar (2)
Öğr. Gör. Ömer Alperen GÜRSES Kırşehir Ahi Evran Üniversitesi, Türkiye
Dr. Öğr. Üyesi İsmail CEYLAN Kırşehir Ahi Evran Üniversitesi, Türkiye
| Makale Türü |
Özgün Makale
(SSCI, AHCI, SCI, SCI-Exp dergilerinde yayınlanan tam makale)
|
||
| Dergi Adı | BMC Medical Education (Q1) | ||
| Dergi ISSN | 1472-6920 Dergi Bilgileri (2026) | ||
| Dergi Tarandığı Indeksler | SSCI | ||
| Makale Dili | Türkçe | Basım Tarihi | 01-2026 |
| Cilt / Sayı / Sayfa | 26 / 1 / – | DOI | 10.1186/s12909-026-08656-3 |
| Makale Linki | https://doi.org/10.1186/s12909-026-08656-3 | ||
| UAK Araştırma Alanları |
Fizyoterapi ve Rehabilitasyon
|
||
| Özet |
| Background: Stability across repeated administrations is essential for educational use of large language models (LLMs), yet it is rarely quantified in non-English, curriculum-aligned anatomy contexts.Methods: Eleven contemporary LLMs answered 100 Turkish, faculty-authored, curriculum-aligned anatomy multiple-choice questions from AYDEP, targeting the undergraduate Physiotherapy and Rehabilitation anatomy curriculum in three independent runs (≥ 12-hour intervals). Testing used developers’ web interfaces in Turkey with browsing disabled and default generation settings (August–September 2025). Performance was summarized with a stability-aware 0–3 item score (number of correct responses across three runs) and predefined response-consistency classes.Results: A subset of models achieved near-ceiling totals with high run-to-run stability, whereas others showed greater session-to-session variability … |
| Anahtar Kelimeler |
| Anatomy examination | ChatGPT | Claude | DeepSeek | Gemini | Grok | Large language models |
BM Sürdürülebilir Kalkınma Amaçları
| Atıf Sayıları | |
| Web of Science | 1 |
| Scopus | 1 |
| Google Scholar | 1 |