Skip to content

Benchmark

Bridging the Gap Winogrande Sepedi (nso) cultural appropriateness target-language minus English lift (QA Pairs with Only “good” Quality Annotations)

Multilingual benchmarks compare model performance across languages.

Models measured
3
Scores recorded
4
Sources
1
IQ weight
No weight published

Score distribution

5.9
Mean
15.6
Median
10
Std dev
-6.9
10th percentile
15.7
90th percentile
-6.9
Min
-6.915.7
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 4osection: Results prose, table: A.19, row: G, statistic: maximum· Bridging the Gap (arXiv:2412.12417v1) data15.6
GPT 4otable: A.19, section: G, language: nso· Bridging the Gap (arXiv:2412.12417v1) data15.7
GPT-3.5 Turbotable: A.19, section: G, language: nso· Bridging the Gap (arXiv:2412.12417v1) data-6.9
GPT-4 Turbotable: A.19, section: G, language: nso· Bridging the Gap (arXiv:2412.12417v1) data-0.8

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)