Gary Marcus

Rank 19 of 47

Score 64

It’s not just the Apple paper. Reasoning outside the distribution has been the core problem for neural networks for 30 years.

Here’s yet another study that is a case in point.

It has been my central goalpost for nearly 30 years, and I will keep emphasizing it until the

Nouha Dziri

@nouhadziri

13d

📢 Can LLMs really reason outside the box in math? Or are they just remixing familiar strategies?

Remember DeepSeek R1, o1 have impressed us on Olympiad-level math but also they were failing at simple arithmetic 😬

We built a benchmark to find out → OMEGA Ω 📐

💥 We found

6/24/2025, 8:27:28 PM

The statement discusses the long-standing challenge of neural networks reasoning outside their training distribution, highlighting a new study as evidence. It emphasizes the importance of this issue in AI research, particularly in mathematical reasoning.

Principle 1:
I will strive to do no harm with my words and actions.
The statement does no harm and focuses on a technical issue in AI research. [+1]
Principle 3:
I will use my words and actions to promote understanding, empathy, and compassion.
It promotes understanding by discussing a significant challenge in AI and referencing a study that addresses it. [+2]
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 by contributing to the discussion on AI capabilities and limitations. [+1]