Gary Marcus

Rank 13 of 47
|
Score 83
5/18/2025, 12:20:38 AM
X
In reply to:
Seán Ó hÉigeartaigh
@S_OhEigeartaigh
·
241d
@GaryMarcus 'Failed' and 'hit significantly diminishing returns' seem different.
Gary Marcus
@GaryMarcus
·
241d
!!!

If this true, it’s compelling evidence that pure pretraining scaling of LLMs has failed, and that that particular approach (which was I specifically said would hit a wall) has failed.
Flowers
@flowersslop
·
241d
I just got feedback from someone who def knows firsthand: GPT-5 will be built from entirely new models that haven’t been released yet. So we’re leaving GPT-4 and all its variants behind and getting a system made from components that aren’t part of anything currently available.

The statement is part of a public discourse on the development and scaling of large language models (LLMs), specifically discussing the potential limitations and future directions of AI models like GPT-5. The conversation involves technical critique and analysis of AI development strategies, which are significant public issues in the field of technology and AI ethics.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The statement does not cause harm; it is a factual correction or clarification in a technical discussion. [+1]
  2. Principle 3:
    I will use my words and actions to promote understanding, empathy, and compassion.
    The statement contributes to understanding by clarifying a point in the discussion, promoting a more accurate view of the situation. [+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 dialogue by addressing a point of disagreement without resorting to personal attacks. [+1]