Skip to content

Benchmark

AfriHG headline generation Arabic (arb) ROUGE-L

Multilingual benchmarks compare model performance across languages. Higher scores are better.

Models and systems measured
2
Scores recorded
2
Sources
1
IQ weight
No weight published

Score distribution

23.7
Mean
24.1
Median
0.45
Std dev
23.2
10th percentile
24.1
90th percentile
23.2
Min
23.224.1
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.

Model or systemScore
AfriTeVa V2 (base)headline generation fine-tunedtable: 1, system: AfriTeVa V2 (base), task: headline generation, language: arb, metric: ROUGE-L· AfriHG (arXiv:2412.20223v1) data24.1
mT5-baseheadline generation fine-tunedtable: 1, system: mT5-base, task: headline generation, language: arb, metric: ROUGE-L· AfriHG (arXiv:2412.20223v1) data23.2

Data credits: Toyib Ogunremi, Serah Akojenu, Anthony Soronnadi, Olubayo Adekanmbi and David Ifeoluwa Adelani. AfriHG: News headline generation for African Languages, arXiv:2412.20223v1. CC BY 4.0 article numeric/protocol extract, restructured with values unchanged. Relayed summarization scores credited to Hasan et al. (2021); no task or model rights asserted. (licence)