Pretraining and Benchmarking Modern Encoders for Latvian
This work provides practical NLP tools for the low-resource Latvian language community, though it is incremental as it applies existing architectures to new data.
The researchers addressed the lack of monolingual Latvian encoders by pretraining a suite of Latvian-specific encoder models based on RoBERTa, DeBERTaV3, and ModernBERT architectures, and evaluating them on Latvian benchmarks. Their best model, lv-deberta-base (111M parameters), achieved the strongest overall performance, outperforming larger multilingual baselines and prior Latvian-specific encoders.
Encoder-only transformers remain essential for practical NLP tasks. While recent advances in multilingual models have improved cross-lingual capabilities, low-resource languages such as Latvian remain underrepresented in pretraining corpora, and few monolingual Latvian encoders currently exist. We address this gap by pretraining a suite of Latvian-specific encoders based on RoBERTa, DeBERTaV3, and ModernBERT architectures, including long-context variants, and evaluating them across a diverse set of Latvian diagnostic and linguistic benchmarks. Our models are competitive with existing monolingual and multilingual encoders while benefiting from recent architectural and efficiency advances. Our best model, lv-deberta-base (111M parameters), achieves the strongest overall performance, outperforming larger multilingual baselines and prior Latvian-specific encoders. We release all pretrained models and evaluation resources to support further research and practical applications in Latvian NLP.