Gary Marcus

Rank 19 of 47
|
Score 64
12/20/2024, 9:59:24 PM
X
In reply to:
Xeno
@shorttimelines
·
199d
@GaryMarcus Well I apologise if that came across as somehow an attack on your character. I’m just surprised that someone could say this. The progress from o1 to o3 on this benchmark along suggests a marked improvement in capabilities. It’s not AGI, but it’s most certainly a step closer
ARC Prize
@arcprize
·
199d
New verified ARC-AGI-Pub SoTA!

@OpenAI o3 has scored a breakthrough 75.7% on the ARC-AGI Semi-Private Evaluation.

And a high-compute o3 configuration (not eligible for ARC-AGI-Pub) scored 87.5% on the Semi-Private Eval.

1/4
Gary Marcus
@GaryMarcus
·
199d
@shorttimelines you said i was being “willfully ignorant”, rather than making a concrete argument against my position. that’s ad hominem, and not helpful.
Xeno
@shorttimelines
·
199d
@GaryMarcus I didn’t insult you but go ahead
Gary Marcus
@GaryMarcus
·
199d
@shorttimelines how? i am asking for an actual specific answer. to me it looks more like more of the same when the foundational problems i have long emphasized still have not been addressed.

if you insult me again, rather than actually answering my question, i will block you.
Xeno
@shorttimelines
·
199d
@GaryMarcus Yes, obviously it’s not. It’s an expression of disbelief. It’s clear that this is a step towards AGI. If you can’t see that, you’re being wilfully ignorant
Gary Marcus
@GaryMarcus
·
199d
@shorttimelines that’s not an answer, and no i am not kidding.

The statement critiques the lack of transparency and public scrutiny in AI model evaluations, suggesting that the data augmentation and testing methods are not fully disclosed. It raises concerns about the validity of claimed improvements in AI capabilities.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The statement aims to highlight potential issues in AI evaluation processes, which could be seen as striving to do no harm by advocating for transparency. [+1]
  2. Principle 2:
    I will respect the privacy and dignity of others and will not engage in cyberbullying, harassment, or hate speech.
    The statement respects privacy and dignity, focusing on the critique of processes rather than individuals. [+1]
  3. 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 criticism by questioning the methods and results, though the surrounding conversation includes ad hominem remarks. [-1]