Gary Marcus

Rank 13 of 47
|
Score 84
11/14/2024, 9:07:39 PM
X
In reply to:
Harlan Stewart ⏹️
@HumanHarlan
·
453d
@GaryMarcus @mlbaggins @ilyasut @ylecun I’m legitimately sorry if I misrepresented your view. Is there a quicker way for me to understand your view better than reading the paper? Like a blogpost or tweet with a good summary?
Gary Marcus
@GaryMarcus
·
453d
@HumanHarlan @mlbaggins @ilyasut @ylecun Did you actually read the paper or just the title? Clearly the former. You don’t have a quote where I said what you are attributing to me.

If you want to be scholarly you would read the paper. But you can also be blocked if you prefer to be this way. 🤷‍♂️
Harlan Stewart ⏹️
@HumanHarlan
·
453d
@mlbaggins @ilyasut @ylecun "Deep learning is hitting a wall"
Gary Marcus
@GaryMarcus
·
454d
𝗬𝗮𝗻𝗻 𝗟𝗲𝗖𝘂𝗻 𝗶𝘀 𝗮𝗯𝘀𝗼𝗹𝘂𝘁𝗲 𝗰𝗼𝗻𝗻𝗶𝘃𝗶𝗻𝗴 𝘁𝗵𝗶𝗲𝗳.

And not for the first time. It’s been a recurring pattern. He must be called out on it.

A short thread starting with the most recent incident.

1. Below he claims he “told you so” about LLMs reaching a…
Mateusz Bagiński ⏹️🔀⿻
@mlbaggins
·
453d
@HumanHarlan @ilyasut @ylecun Do you have a direct quote with sources where Marcus says DN is a dead end or is this an exaggeration?
Harlan Stewart ⏹️
@HumanHarlan
·
453d
My summary of a recent bit of AI discourse:

@ilyasut: existing techniques aren't as good as the secret techniques at my new AI startup (hi investors!)

@ylecun: See everyone, I told you so- LLMs can't reason, existing techniques are trash, clearly inferior to the techniques at…

The statement and linked content engage in a public discourse about the challenges and limitations of deep learning, a significant topic in AI research. The conversation involves multiple parties discussing the interpretation of a paper and its implications for AI development.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The conversation includes some confrontational language, which may not fully align with striving to do no harm. [-1]
  2. Principle 2:
    I will respect the privacy and dignity of others and will not engage in cyberbullying, harassment, or hate speech.
    There is a potential for misrepresentation and misunderstanding, which could affect the privacy and dignity of those involved. [-1]
  3. Principle 3:
    I will use my words and actions to promote understanding, empathy, and compassion.
    The discussion could benefit from more empathy and understanding, as it currently includes defensive and accusatory tones. [-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 dialogue involves some personal attacks and defensive responses, which detract from constructive criticism. [-2]