The evolution of large language models (LLMs) has culminated in a multitask model
paradigm where prompts drive the generation of user-specific outputs. However, this …

Korean is often referred to as a low-resource language in the research community. While
this claim is partially true, it is also because the availability of resources is inadequately …

Korean morphological variations present unique opportunities and challenges in natural
language processing (NLP), necessitating an advanced understanding of morpheme-based …

Training an automatic speech recognition (ASR) post-processor based on sequence-to-
sequence (S2S) requires a parallel pair (eg, speech recognition result and human post …

Visual abductive reasoning is the logical process of drawing the most plausible hypothesis
based on given observations. This ability is fundamental to artificial intelligence because it …

Rather than continuing the conversation based on personalized or implicit information, the
existing conversation system generates dialogue by focusing only on the superficial content …

The evaluation methodology of natural language generation (” NLG”= systems generating
text based on some kind of input) is currently in a state of a hopefully productive confusion …

초록 본 연구는 거대 언어 모델 (Large Language Models, LLM) 의 한국 이해도를 평가하기
위해 설계된 Ko-Sovereign 벤치마크를 제안한다. Ko-Sovereign 벤치마크는 한국어 능력 …

일반 상식 기반의 지식 그래프는 대규모 코퍼스에 포함되어 있는 일반 상식을 그래프로
표현하여, 자연어 처리의 하위 작업들에 적용할 수 있도록 하는 구조화된 지식 표현 방법이다 …

일반 상식 추론 능력은 가장 사람다운 능력 중 하나로써, 인공지능 모델이 쉽게 모사하기 어려운
영역이다. 딥러닝 기반의 언어 모델은 여전히 일반 상식에 기반한 추론을 필요로 하는 분야에서 …