Benchmark
Persian-family Machine Translation BLEU (Persian variant, source-pooled, zero-shot)
Multilingual benchmarks compare model performance across languages. Higher scores are better.
- Models measured
- 2
- Scores recorded
- 2
- Sources
- 1
- IQ weight
- No weight published
Score distribution
- 6.5
- Mean
- 6.7
- Median
- 0.25
- Std dev
- 6.2
- 10th percentile
- 6.7
- 90th percentile
- 6.2
- Min
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Qwen2.5 3B Instructtable: 6, task: Machine Translation, metric: BLEU, subset: Persian variant, source-pooled· Persian, Dari and Tajiki paper data docs | 6.7 |
| 2 | Qwen2.5 1.5B Instructtable: 6, task: Machine Translation, metric: BLEU, subset: Persian variant, source-pooled· Persian, Dari and Tajiki paper data docs | 6.2 |
Data credits: Noor Mairukh Khan Arnob and Abu Bakar Siddique Mahi. One Language, Three of Its Voices, SilkRoadNLP 2026, ©2026 ACL, CC BY 4.0. Numeric/protocol extract from Tables 2, 5 and 6; language and pooled subsets retained. (licence)