BenCzechMark : A Czech-centric Multitask and Multimetric Benchmark for Large Language Models with Duel Scoring Mechanism
This provides a domain-specific benchmark for evaluating large language models on Czech language tasks, addressing a gap for researchers and developers in Czech NLP.
The authors tackled the lack of comprehensive Czech language benchmarks for large language models by creating BenCzechMark, which includes 50 tasks across 8 categories and a new 7B Czech-centric model, achieving baseline results for comparison with multilingual models.
We present BenCzechMark (BCM), the first comprehensive Czech language benchmark designed for large language models, offering diverse tasks, multiple task formats, and multiple evaluation metrics. Its duel scoring system is grounded in statistical significance theory and uses aggregation across tasks inspired by social preference theory. Our benchmark encompasses 50 challenging tasks, with corresponding test datasets, primarily in native Czech, with 14 newly collected ones. These tasks span 8 categories and cover diverse domains, including historical Czech news, essays from pupils or language learners, and spoken word. Furthermore, we collect and clean BUT-Large Czech Collection, the largest publicly available clean Czech language corpus, and use it for (i) contamination analysis and (ii) continuous pretraining of the first Czech-centric 7B language model with Czech-specific tokenization. We use our model as a baseline for comparison with publicly available multilingual models. Lastly, we release and maintain a leaderboard with existing 50 model submissions, where new model submissions can be made at https://huggingface.co/spaces/CZLC/BenCzechMark.