Cross Linguistic
Cross-linguistic research focuses on understanding and leveraging linguistic diversity in computational linguistics, aiming to develop language technologies that work effectively across multiple languages. Current research heavily utilizes large language models (LLMs) and explores techniques like multilingual fine-tuning, parameter-efficient methods (e.g., adapters, LoRA), and cross-lingual transfer learning to address challenges posed by low-resource languages and diverse linguistic structures. This work is crucial for broadening access to NLP tools globally and improving applications such as machine translation, information retrieval, and social media analysis across diverse cultural contexts. Furthermore, investigations into the impact of linguistic typology and cultural factors on model performance are gaining prominence.
Papers
Should We Respect LLMs? A Cross-Lingual Study on the Influence of Prompt Politeness on LLM Performance
Ziqi Yin, Hao Wang, Kaito Horio, Daisuke Kawahara, Satoshi Sekine
J-UniMorph: Japanese Morphological Annotation through the Universal Feature Schema
Kosuke Matsuzaki, Masaya Taniguchi, Kentaro Inui, Keisuke Sakaguchi
Quantifying the Dialect Gap and its Correlates Across Languages
Anjali Kantharuban, Ivan Vulić, Anna Korhonen
Counting the Bugs in ChatGPT's Wugs: A Multilingual Investigation into the Morphological Capabilities of a Large Language Model
Leonie Weissweiler, Valentin Hofmann, Anjali Kantharuban, Anna Cai, Ritam Dutt, Amey Hengle, Anubha Kabra, Atharva Kulkarni, Abhishek Vijayakumar, Haofei Yu, Hinrich Schütze, Kemal Oflazer, David R. Mortensen