Incorporating Syntactic Knowledge into Pre-trained Language Model using Optimization for Overcoming Catastrophic Forgetting
- Ran Iwamoto
- Issei Yoshida
- et al.
- EMNLP 2023
Dr. Hiroshi Kanayama is a Senior Technical Staff Member in the Knowledge Infrastructure Group in IBM Research – Tokyo. He joined IBM just after receiving a Master's Degree in 2000 and worked on natural language processing, mainly syntactic and semantic analysis. His research outcomes on syntactic parser, sentiment analysis and Korean pos tagger were integrated with software products, and they differentiated IBM's text mining solutions. In 2008 he joined the Watson project (Jeopardy! challenge) and provided lexical resources for type matching using information extraction techniques. In 2012 he received a Ph.D. degree from the University of Tokyo with his work on sentiment analysis.