Exploring Multilingual Probing in Large Language Models: A Cross-Language Analysis
Exploring Multilingual Probing in Large Language Models: A Cross-Language Analysis [19.4] 大規模言語モデル(LLM)の探索技術は主に英語に焦点を合わせており、世界の言語の大部分を見下ろしている。 複数のオープンソースのLCMモデルで実験を行い、探索精度、層間の傾向、および複数の言語に対する探索ベクトル間の類似性を解析した。 論文参考訳(メタデータ) (Sun, 22 Sep 2024 14:14:05 GMT)
多言語での動作解析、「(1) a consistent performance gap between high-resource and lowresource languages, with high-resource languages achieving significantly higher probing accuracy; (2) divergent layer-wise accuracy trends, where high-resource languages show substantial improvement in deeper layers similar to English; and (3) higher representational similarities among high-resource languages, with low-resource languages demonstrating lower similarities both among themselves and with high-resource languages.」とのこと