Skip to content

Benchmark

Bridging the Gap Winogrande Sepedi (nso) culturally appropriate subset, target-language evaluation (QA Pairs with at Least One “good” Quality Annotation)

Multilingual benchmarks compare model performance across languages. Scores are percentages. Higher is better.

Models measured
3
Scores recorded
3
Sources
1
IQ weight
No weight published

Score distribution

58.2%
Mean
58.9%
Median
5.7%
Std dev
50.8%
10th percentile
64.8%
90th percentile
50.8%
Min
50.8%64.8%
The band covers the middle 80% of scores. The line marks the median.

Results

The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.

ModelScore
GPT 4otable: A.18, section: A, language: nso· Bridging the Gap (arXiv:2412.12417v1) data64.8%
GPT-3.5 Turbotable: A.18, section: A, language: nso· Bridging the Gap (arXiv:2412.12417v1) data50.8%
GPT-4 Turbotable: A.18, section: A, language: nso· Bridging the Gap (arXiv:2412.12417v1) data58.9%

Data credits: Tuka Alhanai, Adam Kasumovic, Mohammad Ghassemi, Aven Zitzelberger, Jessica Lundin and Guillaume Chabot-Couture. Bridging the Gap: Enhancing LLM Performance for Low-Resource African Languages with New Benchmarks, Fine-Tuning, and Cultural Adjustments, arXiv:2412.12417v1. CC BY 4.0 article numeric/protocol extract, restructured with values unchanged. No task or model rights asserted. (licence)