Large language models can consistently generate high-quality content for election disinformation operations.
大型語言模型可以持續生成高品質內容,用於選舉虛假資訊操作。
PLoS One 2025-03-17
Large language models outperform humans in identifying neuromyths but show sycophantic behavior in applied contexts.
大型語言模型在辨識神經迷思方面優於人類,但在應用情境中表現出諂媚行為。
Trends Neurosci Educ 2025-06-11