Gary Marcus

Rank 11 of 47
|
Score 106
1/6/2025, 4:44:47 AM
X
In reply to:
Gary Marcus
@GaryMarcus
·
205d
5. In o1’s best case, there may have been haeavy data augmentation. In domains where that’s not possible, o1 may not even be that much better than GPT-4 (e.g on some language tasks), as you can see if you read OpenAI’s reports carefully: https://openai.com/index/learnin…
Gary Marcus
@GaryMarcus
·
205d
4. o1 probably isn’t as robust as people might have hoped. Even on certain benchmarks, results may be fragile or difficult to replicate:
Alejandro Cuadron
@Alex_Cuadron
·
205d
Surprising find: OpenAI's O1 - reasoning-high only hit 30% on SWE-Bench Verified - far below their 48.9% claim. Even more interesting: Claude achieves 53% in the same framework. Something's off with O1's "enhanced reasoning"... 🧵1/8
Gary Marcus
@GaryMarcus
·
205d
3. Commonsense reasoning is still dodgy, as it has long been, review here, earlier today: https://garymarcus.substack.com/…
Gary Marcus
@GaryMarcus
·
205d
2. A recent paper on Putnam math problems showed a similar failure on distribution shift, showing 30% decement by o1 on math problems with minor variations eg on variable names.

Same old story, different set of problems.

https://openreview.net/forum?id=…
Gary Marcus
@GaryMarcus
·
205d
1. The problem of distribution shift, well-tested in Apple’s 2024 reasoning paper, which goes back to my own 1998 work, continues.

LLMs, like their predecessors generalize well to similar items but struggle with reliability in unfamiliar regimes, even minor variations sometimes…
Gary Marcus
@GaryMarcus
·
205d
An MIT professor just asked me, how can you be so confident that OpenAI isn’t very close to AGI?

Here is a thread of links to a several recent observations that I think show how far we are from robust intelligence.

Importantly, all of them are *known* problems, many that I…

The statement critiques the robustness and generalization capabilities of AI models, particularly in handling distribution shifts. It highlights the limitations of current models in out-of-domain scenarios, suggesting a need for improvement in AI research.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The statement aims to highlight potential issues in AI model performance, which could lead to improvements, aligning with the principle of doing no harm. [+1]
  2. Principle 3:
    I will use my words and actions to promote understanding, empathy, and compassion.
    By discussing the limitations of AI models, the statement promotes understanding of current technological challenges. [+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 criticism by pointing out specific areas where AI models may fall short, without resorting to personal attacks. [+1]