LLM accuracy does not differ significantly across question languages, ranging from 62% in Polish to 77% in German¶
Claim · All claims
Evidence · 1 study · quant-synthesisr3·q3
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person · recheck by 2028: tool and model accuracy changes quickly
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q3 i? Pooled LLM accuracy ranged from 62% (Polish) to 77% (German) across question languages, and these differences were not statistically significant (P=0.17). → Nouri Haniyeh 2025
Evidence¶
Nouri Haniyeh 2025¶
Nouri Haniyeh, Mahdavi Abdollah, Abedi Ali, Mohammadnia Alireza, Hamedan Mahnaz, Amanzadeh Masoud. (2025). Performance of large language models in medical licensing examinations: a systematic review and meta-analysis. Journal of Educational Evaluation for Health Professions. https://doi.org/10.3352/jeehp.2025.22.36
q3 · i? · quant-synthesis · r3
Subgroup analysis by question language (English n=21, Spanish n=11, Japanese n=10, Chinese n=7, Polish n=6, German n=5, other n=3 studies). The review reports the range and states the differences "were not statistically significant (P=0.17)"; equivalence was not tested.
"In terms of question language, pooled LLM accuracy ranged from 62% to 77%, with the highest accuracy observed for German and the lowest for Polish. These differences, however, were not statistically significant (P=0.17)."
Discussion¶
Related Claims¶
- GPT‐4 through o1 models show significantly lower pooled accuracy in pathology (75%) than in other biomedical disciplines — related
- Under unseen-answers evaluation, the German Transformer model's F-Scores are numerically similar to state-of-the-art Transformer results for English — related
- Large language models achieve a pooled accuracy of 72% on medical licensing examination questions across 36 studies — related
- Translating non-English examination questions into English improved GPT-3.5's performance but had limited effect on GPT-4 — related
- ChatGPT's accuracy on image-based MCQs ranged from 13.1% to 100% across studies, and it performed significantly worse on open-ended questions than on MCQs — related
Page metadata
| Field | Value |
|---|---|
status |
review |
generated |
by process:wiki-ingest on 2026-10-31 |
trust |
unverified — no one has confirmed this page's content |
Frontmatter is the page's machine-readable record — see the Schema & Guide.