Gary Marcus

Rank 18 of 47
|
Score 40
7/24/2024, 10:49:26 PM
X
In reply to:
Burny — Effective Omni
@burny_tech
·
468d
Mainstream LLM benchmarks suck and are full of contamination. This is private noncontaminated reasoning benchmark. You can see how the models are actually getting better, and that were not really "stuck at GPT-4 level intelligence for over a year now".
AI Explained
@AIExplainedYT
·
469d
Llama 3.1 paper is genuinely incredible, inc. near-perfect predictions of benchmark performance from a given compute budget. Most revealing paper of 2024.

Here, though, are the initial results from my SIMPLE bench, as debuted on the channel. 100+ fully private, PhD-vetted,…

The statement constitutes public discourse as it engages in a substantive discussion about the validity and transparency of AI benchmarks, which is a relevant public issue in the field of artificial intelligence. The tone of the statement is critical and challenges the credibility of the information presented by @burny_tech.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The statement does not appear to cause harm directly, but the critical tone could be perceived as dismissive. It could be more constructive in its criticism. Minor violation. [-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, as it does not engage in personal attacks or harassment. Adherence. [+1]
  3. Principle 3:
    I will use my words and actions to promote understanding, empathy, and compassion.
    The statement does not promote understanding, empathy, or compassion. It could be more constructive by suggesting ways to improve the transparency of the benchmark. Minor violation. [-1]
  4. 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 criticism but does not provide constructive feedback or engage in a dialogue. It could be more effective by offering specific suggestions or asking clarifying questions. Minor violation. [-1]