Finetuning LLMs for EvaCun 2025 token prediction shared task
This is an incremental application of existing fine-tuning methods to a new shared task benchmark.
The authors tackled the EvaCun 2025 token prediction task by fine-tuning three large language models (Command-R, Mistral, and Aya Expanse) on provided data without domain-specific adjustments, and they compared three prompting approaches on a held-out dataset.
In this paper, we present our submission for the token prediction task of EvaCun 2025. Our sys-tems are based on LLMs (Command-R, Mistral, and Aya Expanse) fine-tuned on the task data provided by the organizers. As we only pos-sess a very superficial knowledge of the subject field and the languages of the task, we simply used the training data without any task-specific adjustments, preprocessing, or filtering. We compare 3 different approaches (based on 3 different prompts) of obtaining the predictions, and we evaluate them on a held-out part of the data.