個案總覽
依發行單位、學門或關鍵字，找到適合的教學個案。
-
The Working Limitations of Large Language Models
Large language models (LLMs) can generate convincingly human-sounding responses to queries. This ability can lead users to mistakenly attribute certain human capabilities to these artificial intelligence algorithms, namely reasoning, knowledge, understanding, and execution. Understanding how LLMs work and what their limitations are can help users identify where generative AI technology is best applied and where its outputs might be unreliable.