Skip to content

Benchmark

Bridging the Gap Belebele Setswana (tn) human translation back-translated to 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

60.9%
Mean
61.8%
Median
5.6%
Std dev
58.2%
10th percentile
67.7%
90th percentile
50.7%
Min
50.7%67.7%
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 4otable: A.15, section: back-translation-accuracy, task: Belebele, language: tn, shots: 0· Bridging the Gap (arXiv:2412.12417v1) data67.7%
GPT-3.5 Turbotable: A.15, section: back-translation-accuracy, task: Belebele, language: tn, shots: 0· Bridging the Gap (arXiv:2412.12417v1) data60.4%
GPT-4 Turbotable: A.15, section: back-translation-accuracy, task: Belebele, language: tn, shots: 0· Bridging the Gap (arXiv:2412.12417v1) data66.3%
Llama 3 70B Instructtable: A.15, section: back-translation-accuracy, task: Belebele, language: tn, shots: 0· Bridging the Gap (arXiv:2412.12417v1) data61.8%
Llama 3 8B Instructtable: A.15, section: back-translation-accuracy, task: Belebele, language: tn, shots: 0· Bridging the Gap (arXiv:2412.12417v1) data50.7%
Phi-3 Mini 4K Instructtable: A.15, section: back-translation-accuracy, task: Belebele, language: tn, shots: 0· Bridging the Gap (arXiv:2412.12417v1) data58.2%

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)