Conversation

New piece by & me: - language carries a small portion of all human knowledge. - much of human knowledge & all of animal knowledge is non verbal (& non-symbolic). - hence large language models cannot come close to human-level intelligence.
Who is attempting to combine physics models with language models in AI research? Any good resources or papers that you would recommend?
2
6
Better question: who is trying to get a machine to learn a model of the world similar to what a dog or a crow possesses?
7
44
Show replies
Deep understanding by one person is not knowledge. Knowledge is a collective enterprise. As such, knowledge requires communication to be produced, stored, and passed.
Image
1
11
Individuals of non-social animal species have lots of knowledge about the world. They acquire it by themselves without resorting to any form of communication.
3
16
Show replies
Knowledge is acquired by correlating intervention and observation. Action precedes observation and both produce causal knowledge. Free/random action is needed to find causal directions. Agency leads to intelligence, not the other way around. Build robots and set them free.
2
7
A one month old baby hardly has any agency, yet acquires enormous amounts of knowledge about the world, largely by observation.
5
8
Show replies