Upstage, Two Papers Accepted at the World's Leading Conference 'EMNLP 2023'

Upstage announced on the 23rd that it presented two papers at EMNLP 2023, the most prestigious conference in the field of natural language processing (NLP). EMNLP 2023 is a world-class conference that covers research related to language data-based natural language processing approaches such as AI translation, chatbots, and machine reading comprehension.

The most prestigious conference in the field of Natural Language Processing (NLP), EMNLP 2023

Image provided by Upstage

Last year, EMNLP 2022 received a total of 3,242 paper submissions, of which only 715 were accepted, resulting in an acceptance rate of 22%. EMNLP 2023 will be held in Singapore from December 6 to December 10. Leading global AI companies such as Google, Apple, Amazon, and Baidu will participate.

The two papers accepted this time are Korean language-related NLP research achievements, conducted under the leadership of Upstage’s Tech Lead Chanjoon Park in collaboration with Professor Heeseok Lim’s research team at Korea University. The first paper established a new benchmark dataset related to Korean speech recognition post-processing. It proposes a new evaluation methodology to assess and identify weaknesses of speech recognition models. The second paper proposed a new data augmentation technique that leverages the characteristics of the Korean language. It suggests a method for generative language models to diversify Korean morpheme combinations to augment data.

Seonghoon Kim, CEO of Upstage, said, "We are very pleased to achieve research results at various global conferences, including EMNLP 2023," and added, "Based on research outcomes through continuous research and development (R&D) investment, we will do our best to enable everyone to conveniently use AI with the highest performance."

This content was produced with the assistance of AI translation services.