[recitedropper]

Perception seems to be one of the main constraints on LLMs that not much progress has been made on. Perhaps not surprising, given perception is something evolution has worked on since the inception of life itself. Likely much, much more expensive computationally than it receives credit for.

[Workaccount2]

I strongly suspect it's a tokenization problem. Text and symbols fit nicely in tokens, but having something like a single "dog leg" token is a tough problem to solve.

[stalfie]

The neural network in the retina actually pre-processes visual information into something akin to "tokens". Basic shapes that are probably somewhat evolutionarily preserved. I wonder if we could somehow mimic those for tokenization purposes. Most likely there's someone out there already trying.

(Source: "The mind is flat" by Nick Chater)

[recitedropper]

I think in this case, tokenization and percpetion are somewhat analogous. I think it is probably the case our current tokenization schemes are really simplistic compared to what nature is working with. If you allow the analogy.

[orly01]

Why should it have to be expensive computationally? How do brains do it with such a low amount of energy? I think catching the brain abilities even of a bug might be very hard, but that does not mean that there isn't a way to do it with little computational power. It requires having the correct structures/models/algorithms or whatever is the precise jargon.

[nomel]

> How do brains do it with such a low amount of energy?

Physical analog chemical circuits whose physical structure directly is the network, and use chemistry/physics directly for the computations. For example, a sum is usually represented as the number of physical ions present within a space, not some ALU that takes in two binary numbers, each with some large number of bits, requiring shifting electrons to and from buckets, with a bunch of clocked logic operations.

There are a few companies working on more "direct" implementations of inference, like Etched AI [1] and IBM [2], for massive power savings.

[1] https://en.wikipedia.org/wiki/Etched_(company)

[2] https://spectrum.ieee.org/neuromorphic-computing-ibm-northpo...

[recitedropper]

This is the million dollar question. I'm not qualified to answer it, and I don't really think anyone out there has the answer yet.

My armchair take would be that watt usage probably isn't a good proxy for computational complexity in biological systems. A good piece of evidence for this is from the C. elegans research that has found that the configuration of ions within a neuron--not just the electrical charge on the membrane--record computationally-relevant information about a stimulus. There are probably many more hacks like this that allow the brain to handle enormous complexity without it showing up in our measurements of its power consumption.