Benchmark
FormosanBench summarization ROUGE-2 (Atayal, zero-shot)
Multilingual benchmarks compare model performance across languages. Scores are percentages. Higher is better.
- Models measured
- 2
- Scores recorded
- 2
- Sources
- 1
- IQ weight
- No weight published
Score distribution
- 7%
- Mean
- 11.1%
- Median
- 4.1%
- Std dev
- 2.9%
- 10th percentile
- 11.1%
- 90th percentile
- 2.9%
- Min
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | mistral-7b-v0.3zero-shot, Mandarin prompt· FormosanBench (arXiv:2506.21563v1) data | 11.1% |
| 2 | GPT 4ozero-shot, Mandarin prompt· FormosanBench (arXiv:2506.21563v1) data | 2.9% |
Data credits: Lin, Chen and Zhang, FormosanBench: Benchmarking Low-Resource Austronesian Languages in the Era of Large Language Models, arXiv:2506.21563v1, CC BY 4.0. Scores extracted from Tables 2-4. (licence)