Skip to content

Benchmark

Bridging the Gap Belebele Zulu (zu) human-translation minus machine-translation difference

Multilingual benchmarks compare model performance across languages.

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

Score distribution

-3.7
Mean
-3.7
Median
2.6
Std dev
-6.3
10th percentile
0.7
90th percentile
-6.3
Min
-6.30.7
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: machine-translation-difference, task: Belebele, language: zu, shots: 0· Bridging the Gap (arXiv:2412.12417v1) data-6.3
GPT-3.5 Turbotable: A.15, section: machine-translation-difference, task: Belebele, language: zu, shots: 0· Bridging the Gap (arXiv:2412.12417v1) data-1.3
GPT-4 Turbotable: A.15, section: machine-translation-difference, task: Belebele, language: zu, shots: 0· Bridging the Gap (arXiv:2412.12417v1) data-6.3
Llama 3 70B Instructtable: A.15, section: machine-translation-difference, task: Belebele, language: zu, shots: 0· Bridging the Gap (arXiv:2412.12417v1) data-5.3
Llama 3 8B Instructtable: A.15, section: machine-translation-difference, task: Belebele, language: zu, shots: 0· Bridging the Gap (arXiv:2412.12417v1) data-3.7
Phi-3 Mini 4K Instructtable: A.15, section: machine-translation-difference, task: Belebele, language: zu, shots: 0· Bridging the Gap (arXiv:2412.12417v1) data0.7

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)