Adapting AlignScore Mertic for Factual Consistency Evaluation of Text in Russian: A Student Abstract
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914185176154112 |
|---|---|
| author | Zimin, Mikhail Shamsutdinova, Milyausha Andriushchenko, Georgii |
| author_facet | Zimin, Mikhail Shamsutdinova, Milyausha Andriushchenko, Georgii |
| contents | Ensuring factual consistency in generated text is crucial for reliable natural language processing applications. However, there is a lack of evaluation tools for factual consistency in Russian texts, as existing tools primarily focus on English corpora. To bridge this gap, we introduce AlignRuScore, a comprehensive adaptation of the AlignScore metric for Russian. To adapt the metric, we fine-tuned a RuBERT-based alignment model with task-specific classification and regression heads on Russian and translated English datasets. Our results demonstrate that a unified alignment metric can be successfully ported to Russian, laying the groundwork for robust multilingual factual consistency evaluation. We release the translated corpora, model checkpoints, and code to support further research. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_06586 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Adapting AlignScore Mertic for Factual Consistency Evaluation of Text in Russian: A Student Abstract Zimin, Mikhail Shamsutdinova, Milyausha Andriushchenko, Georgii Computation and Language Ensuring factual consistency in generated text is crucial for reliable natural language processing applications. However, there is a lack of evaluation tools for factual consistency in Russian texts, as existing tools primarily focus on English corpora. To bridge this gap, we introduce AlignRuScore, a comprehensive adaptation of the AlignScore metric for Russian. To adapt the metric, we fine-tuned a RuBERT-based alignment model with task-specific classification and regression heads on Russian and translated English datasets. Our results demonstrate that a unified alignment metric can be successfully ported to Russian, laying the groundwork for robust multilingual factual consistency evaluation. We release the translated corpora, model checkpoints, and code to support further research. |
| title | Adapting AlignScore Mertic for Factual Consistency Evaluation of Text in Russian: A Student Abstract |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2512.06586 |