Hacker News new | ask | show | jobs
by RohMin 1179 days ago
The reason I ask is because I feel that a memory model is one of the major bottlenecks toward AGI.
2 comments

On a more serious note, I do agree with you that memory and self-excitation seem like they are the last push thats needed to get to something more akin to "AGI". But I don't think that Rubicon will be crossed with plugins.
>I do agree with you that memory and self-excitation seem like they are the last push thats needed to get to something more akin to "AGI"

"We show that transformer-based large language models are computationally universal when augmented with an external memory. Any deterministic language model that conditions on strings of bounded length is equivalent to a finite automaton, hence computationally limited. However, augmenting such models with a read-write memory creates the possibility of processing arbitrarily large inputs and, potentially, simulating any algorithm."

From "Memory Augmented Large Language Models are Computationally Universal"

https://deepai.org/publication/memory-augmented-large-langua...

why? short and long-term memory is really easy to do. Even my own basic assistant has it (running on fine-tuned curie model)