cafe

A Watermark for Large Language Models (2023) 2301.10226v4.pdf 1.39MB [초록] Potential harms of large language models can be mitigated by watermarking model output, i.e., embedding signals into generated text that are invisible to humans...

Lee, J. H., Shin, D, & Hwang, Y. (2024). Investigating the capabilities of large language model-based task-oriented dialogue chatbots from a learner’s perspective. System, 127, 103538. (94번째 Journal Paper) https://doi.org/10.1016/j...

수행하도록 하는 인공지능 모델 기본적으로 멀티모달(비전+언어) 모델 -> encoder : ViT(vision transformer) + LLM(large language model) -> decoder : action token(command) 생성 바퀴형로봇(자동차), 로봇팔(매니퓰레이터), 인간형로봇(휴머노이드)에...

universities and Lotte Innovate, and is pursuing the development of a proprietary 200-billion-parameter large language foundation model package with world-class performance. NC AI 팀은 여러 국내 대학과 롯데이노베이트로 구성되어 있으며...