Benchmark
FormosanBench MT BLEU (Atayal→Mandarin, zero-shot)
Multilingual benchmarks compare model performance across languages. Higher scores are better.
- Models measured
- 5
- Scores recorded
- 5
- Sources
- 1
- IQ weight
- No weight published
Score distribution
- 0
- Mean
- 0
- Median
- 0
- Std dev
- 0
- 10th percentile
- 0
- 90th percentile
- 0
- Min
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Gemma 1.1 7B ITzero-shot, Mandarin prompt· FormosanBench (arXiv:2506.21563v1) data | 0 |
| 2 | GPT-4o minizero-shot, Mandarin prompt· FormosanBench (arXiv:2506.21563v1) data | 0 |
| 3 | Llama 3.1 70Bzero-shot, Mandarin prompt· FormosanBench (arXiv:2506.21563v1) data | 0 |
| 4 | Llama 3.1 8Bzero-shot, Mandarin prompt· FormosanBench (arXiv:2506.21563v1) data | 0 |
| 5 | Mistral (7B) Instruct v0.3zero-shot, Mandarin prompt· FormosanBench (arXiv:2506.21563v1) data | 0 |
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)