Refining Czech GEC: Insights from a Multi-Experiment Approach

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Pechman, Petr, Straka, Milan, Straková, Jana, Náplava, Jakub
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912556436684800
author Pechman, Petr
Straka, Milan
Straková, Jana
Náplava, Jakub
author_facet Pechman, Petr
Straka, Milan
Straková, Jana
Náplava, Jakub
contents We present a grammar error correction (GEC) system that achieves state of the art for the Czech language. Our system is based on a neural network translation approach with the Transformer architecture, and its key feature is its real-time synthetic generation pipeline, which dynamically augments sentences with artificial errors by introducing both language-agnostic and Czech-specific errors. We conduct a comprehensive series of experiments, investigating the Czech GEC corpora as bases for synthetic error introduction, several error generation strategies, domain balancing, tokenization granularity, model size, and data scaling during fine-tuning. Additionally, we evaluate the performance of large language models (LLMs) on Czech GEC in both end-user and expert fine-tuning scenarios. Our best-performing model is superior both in performance and computational efficiency. The source code and the trained model links are available on https://github.com/ufal/tsd2025-gec.
format Preprint
id arxiv_https___arxiv_org_abs_2506_22402
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Refining Czech GEC: Insights from a Multi-Experiment Approach
Pechman, Petr
Straka, Milan
Straková, Jana
Náplava, Jakub
Computation and Language
We present a grammar error correction (GEC) system that achieves state of the art for the Czech language. Our system is based on a neural network translation approach with the Transformer architecture, and its key feature is its real-time synthetic generation pipeline, which dynamically augments sentences with artificial errors by introducing both language-agnostic and Czech-specific errors. We conduct a comprehensive series of experiments, investigating the Czech GEC corpora as bases for synthetic error introduction, several error generation strategies, domain balancing, tokenization granularity, model size, and data scaling during fine-tuning. Additionally, we evaluate the performance of large language models (LLMs) on Czech GEC in both end-user and expert fine-tuning scenarios. Our best-performing model is superior both in performance and computational efficiency. The source code and the trained model links are available on https://github.com/ufal/tsd2025-gec.
title Refining Czech GEC: Insights from a Multi-Experiment Approach
topic Computation and Language
url https://arxiv.org/abs/2506.22402