Gary Marcus

Rank 9 of 47
|
Score 121
Gorgi Kosev
@spion
·
45d
@VraserX @GaryMarcus @shaunralston How do we know whether LLMs measure better because they are more capable, or because the more commonly explained problems and benchmarks end up suffering from training data contamination in the next training batch?
6/15/2025, 3:16:09 PM
X

The statement 'We don’t' is a brief response in a complex discussion about the capabilities and evaluation of large language models (LLMs). The conversation involves critiques and defenses of AI progress and predictions. The statement itself is vague and does not contribute substantively to the discourse.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The statement neither harms nor benefits the discourse as it lacks substance.
  2. Principle 2:
    I will respect the privacy and dignity of others and will not engage in cyberbullying, harassment, or hate speech.
    It does not engage in harassment or disrespect, but it also does not promote respect or dignity.
  3. Principle 3:
    I will use my words and actions to promote understanding, empathy, and compassion.
    The statement does not promote understanding or empathy due to its brevity and lack of clarity.
  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.
    It does not engage in constructive criticism or dialogue, as it provides no argument or evidence. [-1]