Multicultural and Multimodal Natural Language Processing (National Institute of Information and Communications Technology (NICT))
Research Staff
-
Professor
Masao UTIYAMA -
Associate Professor
Jin TEI
| mmnlp.naist@ml.nict.go.jp | |
| To the site | https://astrec.nict.go.jp/naist/ |
Research Areas
The main research directions are listed below. Please feel free to consult with us as various research subjects are possible, not limited to those listed below.
Construction of multilingual data
We construct training data necessary for multicultural processing based on multilingual data. This involves not only languages for which large amounts of data are already available, such as English and Japanese, but also focusing on developing data that can reflect the numerous under-studied languages and their cultures.
Generation of multicultural information
Generating human-readable and easy-to-understand text and images using computers is considered one of the most important aspects of information processing. Our research emphasizes addressing cultural diversity, robustness to input variations, and the explainability of the generated results.
Multimodal information processing
We are also working on the integration of image, audio, and text information, as well as processing that utilizes this information.
Key Features
In today's world, where AI technology continues to evolve based on the increasing scale of data and computation, researchers and developers who possess both a multicultural sensibility and the ability to integrate and process information in different formats are essential. We aim to cultivate students from diverse fields according to their individual knowledge.
Internships at NICT are also available, offering practical experience in research settings that consider the social implementation of AI technology, and fostering a broader perspective through discussions with diverse stakeholders.