Gary Marcus直接反驳'有效就是神经符号'的说法，坚持按符号操作来定义。搞不清神经符号是啥的看这条。
Gary Marcus在X上回应Florian Tramèr的推文，指出神经符号AI的定义并未改变。他认为只要同时使用条件判断、变量操作、代码解释器等符号操作和神经网络，就属于神经符号系统。Marcus强调自2001年以来他一直在阐述这一定义，只是近年来的实践发生了变化。他宣称过去三年神经符号方法已经获胜。
Wrong! If an AI leverages symbolic operations (conditionals, operations over variables, code interp...
Wrong! If an AI leverages symbolic operations (conditionals, operations over variables, code interpreters, etc) and neural networks it is neurosymbolic. If it doesn’t, it’s not. ( I laid all of this out in 2001 and dozens of articles since then.) What has changed is NOT the definitions, but what people are doing. Over the last three years neurosymbolic has won, plain and simple, fair and square. Florian Tramèr @florian_tramer If it works, it's "neurosymbolic" If it doesn't, it's "plain LLMs" 🔗 View Quoted Tweet 💬 4 🔄 1 ❤️ 17 👀 3682 📊 5 ⚡