aya-definition-ru-axolotl24st

This model is a version of CohereLabs/aya-101, fine-tuned on datasets of Russian usage examples and definitions.

It generates definitions of Russian words in context. Its input is the usage example and the instruction question "Что такое ?"

Other models

Finnish

• decoder-only

Tower, axolotl24 Tower, axolotl24 + dbnary

• encoder-only

mT0-xl, axolotl24 mT0-xl, axolotl24 + dbnary

aya-101, axolotl24 aya-101, axolotl24 + dbnary

German

• decoder-only

Tower, dbnary

• encoder-only

mT0-xl, dbnary

aya-101, dbnary

Russian

• decoder-only

Tower, axolotl24 Tower, axolotl24 + dbnary

• encoder-only

mT0-xl, axolotl24 mT0-xl, axolotl24 + dbnary

aya-101, axolotl24 aya-101, axolotl24 + dbnary

Model Sources

• Repository: MultilingualDefGen
• Paper: accepted to EMNLP 2025 Findings

Uses

The model is intended for research purposes, as a source of contextualized dictionary-like lexical definitions.

The fine-tuning datasets were limited to Russian. Although the original model is multilingual, we did not evaluate its ability to generate definitions in other languages.

Generated definitions can contain all sorts of biases and stereotypes, stemming from the underlying language model and raw dictionary data.

Direct Use

script to run prediction

Training Details

Training Data

axolotl24

Training Procedure

script to run training

Evaluation

run evaluation

Testing Data

axolotl24 Russian test set

Metrics

BLEU, BERTScore

Citation

BibTeX:

@misc{fedorova2025explainingnovelsensesusing,
      title={Explaining novel senses using definition generation with open language models}, 
      author={Mariia Fedorova and Andrey Kutuzov and Francesco Periti and Yves Scherrer},
      year={2025},
      eprint={2509.26181},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2509.26181}, 
}

Framework versions

bert-score==0.3.13
peft==0.14.0
sentencepiece==0.2.0
tokenizers==0.20.1
torch==2.2.2
transformers==4.46.1
trl==0.15.2