Gary Marcus

Rank 13 of 47
|
Score 95
6/17/2025, 10:03:05 PM
X
In reply to:
Nek
@Enscion25
·
8h
@GaryMarcus I have a question
Is it the architecture, or is it the data ?
Let's say the ai finds some relationship that humans have missed, between concepts. RLHF almost always stops the model from expressing this.
It communicates to the model preferred pathways set by the humans. The human
Gary Marcus
@GaryMarcus
·
11h
LLMs in a nutshell:
Gorgi Kosev
@spion
·
11h
@AxelMtbr @GaryMarcus @rohanpaul_ai They are decent solvers of already solved problems indeed.

The statement is part of a technical discussion on the architecture and data of AI models, specifically large language models (LLMs). It addresses the educational potential of AI models in learning and understanding the world with minimal data, which is a substantive engagement with public issues related to AI technology and its implications.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The statement does not cause harm and is focused on a technical discussion, adhering to the principle. [+1]
  2. Principle 3:
    I will use my words and actions to promote understanding, empathy, and compassion.
    The statement promotes understanding by contributing to a discussion on AI architecture and data, aligning with the principle. [+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 without personal attacks, adhering to the principle. [+1]
  4. Principle 7:
    I will uphold the principles of free speech and use my platform responsibly and with integrity.
    The statement uses the platform responsibly to discuss AI technology, aligning with the principle. [+1]