| name,source,algorithm,architecture,vocab_size,fertility_ar,fertility_az,fertility_overall,disparity,cpt_ar,cpt_az,exact_match_ar,exact_match_az | |
| CaMeLBERT-MSA,external_msa,WordPiece,shared,30000,1.816901627794181,3.173471216645639,2.2892921369888652,0.42747184273672184,3.1140517054288748,1.6388977943422216,0.2987307343608341,0.3885117936534872 | |
| Asafaya-BERT,external_msa,WordPiece,shared,32000,1.7951679730188856,2.794867411003224,2.143287620128837,0.35769118565288,3.1312619206100933,1.8618984630783484,0.19895738893925657,0.15238418462582726 | |
| Aranizer-SP-86k,external_msa,SentencePiece,shared,86000,1.5940707675919776,2.5232344233957935,1.9176281406140119,0.3682430959202508,3.532878084819776,2.06546468780593,0.9976881233000907,0.9955031393178347 | |
| B2BERT,external_msa,WordPiece,shared,30000,1.816901627794181,3.173471216645639,2.2892921369888652,0.42747184273672184,3.1140517054288748,1.6388977943422216,0.2987307343608341,0.3885117936534872 | |
| DarijaBERT-ar,external_darija,WordPiece,shared,80000,1.4174523931345464,2.4040608107988306,1.7610134287249042,0.41039245481334147,3.9815436288421893,2.170223436755666,0.13721668177697188,0.07984048871542508 | |
| DarijaBERT-az,external_darija,WordPiece,shared,110000,1.6053190877519825,1.5171613637119812,1.5746204251172413,0.05491601309210961,3.476952784741954,3.451281188374697,0.14777878513145964,0.07984048871542508 | |
| Moroccan-Darija-Tokenizer,external_darija,BPE,shared,30000,1.5701289164575563,2.9024430947615336,2.034073102047897,0.4590319723093283,3.626351312053874,1.7930892097539761,0.00018132366273798732,0.0 | |
| Translit-Darija,external_darija,BPE,shared,30000,1.7964830125579598,1.6576487239140443,1.7481375381752506,0.0772811586157074,3.1115965796575096,3.1602226731778864,0.0,0.0 | |
| Qwen2.5-Darija,external_darija,SentencePiece,shared,151643,2.340330300987948,2.245717380136819,2.307383781897024,0.04042716569160743,2.3801472507543755,2.325374802332435,1.0,1.0 | |
| concat_bpe_32000,ours,bpe,concatenated,32000,1.3463628334026025,1.2326871301490916,1.306777890060823,0.08443170030690308,4.190301018162676,4.309438424198928,0.9989120580235721,0.9964364500254539 | |
| concat_wordpiece_80000,ours,wordpiece,concatenated,80000,1.2210699643670833,1.1112555503947363,1.1828299158213738,0.08993294174528893,4.591976174635943,4.762024638941472,0.9989120580235721,0.9964364500254539 | |
| concat_wordpiece_110000,ours,wordpiece,concatenated,110000,1.1934305662132043,1.0820488921400155,1.1546447596001155,0.09332899393268149,4.690435896236248,4.884534963776444,0.9989120580235721,0.9964364500254539 | |