Yes, there is an inverse correlation between task length and model performance — because model reasoning abilities decrease dramatically with context size, as has been demonstrated in many different domains with sequence-only data.
RESEARCH
-
ARC-AGI relies on symbolic reasoning, not visual perception
By
–
If ARC-AGI required visual perception, you'd see VLMs outperform strict LLMs — by a lot. Everyone tried VLMs during the 2024 competition — no one got better results. Every single top entry used a strict LLM. As we've said many times: ARC-AGI is a 2D symbolic reasoning
-

The Tensor Cookbook: Essential Resource for Machine Learning
By
–
The Tensor Cookbook https://
bit.ly/3zx8UQx
#AI #MachineLearning #DeepLearning #LLMs #DataScience -

o1-preview shows promise for medical reasoning and healthcare integration
By
–
promising results of o1-preview for medical reasoning (though still lots of work to figure out how to integrate with the healthcare system)
-
Author Notices ChatGPT Echoes Writing Habits from 2014 Book
By
–
Haha fun fact: I used to”delve” 2x in my 2014 Python Machine Learning book 10 years as well. ChatGPT must have learned that habit from my older books
-
Deep Learning Origins Traced to Canadian Contributions
By
–
All of deep learning was imported from Canada
-
AI May Solve Math Problems Before Mastering Animal Robotics
By
–
It’s entirely likely that computers will figure out how to solve the biggest open problems in mathematics (eg the RH) before robotics control software figures how to faithfully replicate a cat/dog.
-

2024 Synthetic Data and Small Models: Best Papers Recap
By
–
## Recapping 2024 in Synthetic Data/Smol Models! We are very honored to have @LoubnaBenAllal1 recap and pick all the best papers in: Synthetic Data and Smol Models this year! Timestamps [00:00:05] Loubna Intro
[00:00:33] The Rise of Synthetic Data Everywhere
[00:02:57] -
All Technology Can Be Threatening: Need for Thoughtful Progress
By
–
All tech can be threating, we need to be thoughtful as we continue to progress.
-
ARC Benchmark Limitations: Smoke Tests vs Real AI Progress
By
–
that's largely been my opinion on beating ARC. Over the past decade, I've seen too many artificial benchmarks becoming the target, and being beaten, and resulted in no or little meaningful actual progress.
I guess it can be a smoke test, but I don't think it's going to be a