Gary Marcus

Rank 13 of 47
|
Score 83
5/18/2025, 1:23:30 PM
X
In reply to:
David Manheim
@davidmanheim
·
247d
@GaryMarcus @uhbif19 @XRobservatory It's not a test of the world model because image generation is using diffusion to generate something in the latent space of images it was trained on, not something in the language model.

It's like saying hand-eye coordination is part of understanding; it's kind-of true, but, no.
Gary Marcus
@GaryMarcus
·
247d
@uhbif19 @davidmanheim @XRobservatory I documented a multiple similars one for ChatGPT (not o3) in text but curious why everyone thinks image generation is not a legitimate test of a systems “world model”
There is no other room 🇬🇪
@uhbif19
·
247d
@GaryMarcus @davidmanheim @XRobservatory But it does not make errors like that in text. It just have bad image comprehension, that is all.
Gary Marcus
@GaryMarcus
·
247d
@davidmanheim @uhbif19 @XRobservatory What’s your theory of the psychology of a system that can make an error like that?
David Manheim
@davidmanheim
·
247d
@uhbif19 @XRobservatory @GaryMarcus was initially claiming that "LLMs that don’t understand clocks and time zones." I agree that image generation is not yet at a point where it gets this right, and that's a real flaw in the models, but the original claim as stated is false.
There is no other room 🇬🇪
@uhbif19
·
247d
@davidmanheim @XRobservatory But that is code, the issue is with the direct image output.

The statement is part of a technical discussion about the capabilities and limitations of AI models, specifically regarding image generation and world models. It seeks clarification on a point made in the ongoing conversation, indicating a desire to understand the technical nuances better.

  1. Principle 3:
    I will use my words and actions to promote understanding, empathy, and compassion.
    The statement promotes understanding by seeking clarification on a technical point, which can lead to a more informed discussion. [+1]
  2. Principle 4:
    I will engage in constructive criticism and dialogue with those in disagreement and will not engage in personal attacks or ad hominem arguments.
    The statement engages in constructive dialogue by asking a clarifying question without resorting to personal attacks or dismissive language. [+1]