Skip to content

Benchmark

Bridging the Gap Belebele Xhosa (xh) machine-translated from English

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

Models measured
6
Scores recorded
6
Sources
1
IQ weight
No weight published

Score distribution

50.9%
Mean
38.7%
Median
24.2%
Std dev
32.9%
10th percentile
86.8%
90th percentile
28%
Min
28%86.8%
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.15, section: machine-translation-accuracy, task: Belebele, language: xh, shots: 0· Bridging the Gap (arXiv:2412.12417v1) data86.8%
GPT-3.5 Turbotable: A.15, section: machine-translation-accuracy, task: Belebele, language: xh, shots: 0· Bridging the Gap (arXiv:2412.12417v1) data32.9%
GPT-4 Turbotable: A.15, section: machine-translation-accuracy, task: Belebele, language: xh, shots: 0· Bridging the Gap (arXiv:2412.12417v1) data82.8%
Llama 3 70B Instructtable: A.15, section: machine-translation-accuracy, task: Belebele, language: xh, shots: 0· Bridging the Gap (arXiv:2412.12417v1) data38.7%
Llama 3 8B Instructtable: A.15, section: machine-translation-accuracy, task: Belebele, language: xh, shots: 0· Bridging the Gap (arXiv:2412.12417v1) data36.1%
Phi-3 Mini 4K Instructtable: A.15, section: machine-translation-accuracy, task: Belebele, language: xh, shots: 0· Bridging the Gap (arXiv:2412.12417v1) data28%

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)