扩充希腊方言语料库，支持多模型跨架构评估
GRDD+: An Extended Greek Dialectal Dataset with Cross-Architecture Fine-tuning Evaluation
- 新增6种希腊方言，总语料达637万词，涵盖10种方言
- 在3种Llama系列与Krikri模型上验证方言数据提升效果
- 首次系统对比主流大模型在方言任务中的表现
我们提出了扩展版希腊方言语料库GRDD+，在原有GRDD基础上新增克里特、塞浦路斯、庞蒂克及北希腊方言数据，并新增六种方言：科西嘉希腊语、伊特鲁里亚希腊语（Griko）、曼尼奥特语、赫普塔内西亚语、塔萨孔语和卡瑟雷沃萨希腊语。最终语料库总规模达6,374,939词，包含10种方言，是迄今规模最大、多样性最高的希腊方言数据集。我们对三种模型架构（Llama-3-8B、Llama-3.1-8B、Krikri-8B）进行微调实验，评估高质量方言数据对大语言模型的影响，并与前沿模型（Claude-3.7-Sonnet、Gemini-2.5、ChatGPT-5）进行对比。
原文摘要 · Abstract (English)
We present an extended Greek Dialectal Dataset (GRDD+) 1that complements the existing GRDD dataset with more data from Cretan, Cypriot, Pontic and Northern Greek, while we add six new varieties: Greco-Corsican, Griko (Southern Italian Greek), Maniot, Heptanesian, Tsakonian, and Katharevusa Greek. The result is a dataset with total size 6,374,939 words and 10 varieties. This is the first dataset with such variation and size to date. We conduct a number of fine-tuning experiments to see the effect of good quality dialectal data on a number of LLMs. We fine-tune three model architectures (Llama-3-8B, Llama-3.1-8B, Krikri-8B) and compare the results to frontier models (Claude-3.7-Sonnet, Gemini-2.5, ChatGPT-5).
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。