AuraTracer智迹闻
中文

EVENT DOSSIER

MoirfEolas and Cr\'iochScore: Developing Resources for and the Evaluation of Tokenization Alignment with Irish Morphology

2026-09-07 12:00 Science across 2 days 🔥 47.2 heat score
2sources
2days unfolding
47.2heat score
3mentions
SummaryAI generated

On September 4, 2026, arXiv cs.CL released the MoirfEolas dataset and CríochScore evaluation metrics. MoirfEolas contains over 35,000 Irish words along with their root, prefix, and suffix mappings, aimed at improving the low-resource status of the language and serving as a model for other languages. CríochScore is used to measure the consistency of word segmentation algorithms with morphological boundaries. The study utilized this metric to evaluate common word segmentation algorithms and found that word language models more often maintained morphological consistency with Irish language when compared to other algorithms. Additionally, the study revealed the trade-offs between morphological alignment, compression rate, and lexical efficiency, providing practical insights for the development of Irish natural language processing.

Related eventsRELATED EVENTS
Key entitiesKEY ENTITIES
Cr'iochScoreCríochScoreMoirfEolas

Coverage · reports per dayLANGUAGE SPLIT

Entity relations
Cr'iochScore × MoirfEol…1CríochScore × MoirfEolas1

Integrated timelineUNIFIED TIMELINE

  1. 2026-09-04

    MoirfEolas and CríochScore: Developing …

    本研究发布了针对爱尔兰语的新分词资源及与形态边界对齐的评估指标。作者提出了包含超过 35,000 个爱尔兰单词及其对应前缀、后缀和词根映射数据的 MoirfEolas 数据集,以及用于评估分词与 MoirfEolas 中形态边界对齐程度的…

  2. 2026-09-07

    MoirfEolas and Cr\'iochScore: Developin…

    本文发布 MoirfEolas 数据集及 Cr\'iochScore 评估指标,旨在开发爱尔兰语分词资源并评估其与形态边界的一致性。MoirfEolas 包含超过 35,000 个爱尔兰单词及其对应的词根、前缀和后缀映射;Cr\'ioch…

SignalsSIGNALS

Keyword heat
  • MoirfEolas2
  • Cr'iochScore1
  • CríochScore1

All reports (2)SOURCES

A arXiv cs.CL en 2026-09-04 19:41

MoirfEolas and CríochScore: Developing Resources for and the Evaluation of Tokenization Alignment with Irish Morphology

本研究发布了针对爱尔兰语的新分词资源及与形态边界对齐的评估指标。作者提出了包含超过 35,000 个爱尔兰单词及其对应前缀、后缀和词根映射数据的 MoirfEolas 数据集,以及用于评估分词与 MoirfEolas 中形态边界对齐程度的 CríochScore 指标。研究利用上述指标及现有分词文献中的内在指标,对常用分词算法进行了评估,发现单词语言模型在与其他算法相比时更常与爱尔兰语形态保持一致。此外,研究揭示了分词在形态对齐、压缩效率及词汇效率之间的权衡关系,为爱尔兰自然语言处理开发提供了实践见解。该数据集有助于缓解爱尔兰语的低资源状态,其构建过程可供其他语言仿效以创建专门的形态资源。

A arXiv cs.CL en 2026-09-07 12:00

MoirfEolas and Cr\'iochScore: Developing Resources for and the Evaluation of Tokenization Alignment with Irish Morphology

本文发布 MoirfEolas 数据集及 Cr\'iochScore 评估指标,旨在开发爱尔兰语分词资源并评估其与形态边界的一致性。MoirfEolas 包含超过 35,000 个爱尔兰单词及其对应的词根、前缀和后缀映射;Cr\'iochScore 用于衡量分词与 MoirfEolas 中形态边界的对齐程度。研究利用 Cr\'iochScore 及文献中的内在指标评估了常用分词算法,发现单词语言模型在与其他算法相比时更常与爱尔兰语形态保持一致。此外,研究揭示了分词形态对齐与压缩率及词汇效率之间的权衡关系,为爱尔兰自然语言处理开发提供实践见解。该数据集有助于缓解爱尔兰语的低资源状态,其构建过程可供其他语言仿制以创建专门的形态资源。