Pinned
Surprise of the past 4 years: memory is a harder problem than intelligence.
Jack Cole
2,150 posts
Jack Cole
@MindsAI_Jack
Independent AI Researcher, PhD in Clin Psych, App Dev
40M+ App Downloads, Competive ML (Kaggle Expert)
Seeking a compute sponsor for 2026 - Contact me
- ARC-AGI V2 is quite challenging. We're happy to be back at the top at 12.36! @MohamedOsmanML @bayesilicon @tufalabs @arcprize
- @arcprize 2024 with more than 16k entrants just ended after 5 months, and we rank #1 (@bayesilicon @MohamedOsmanML)! We just scored 58% with a submission that finished after the deadline! We're just getting started. We hope to have an announcement about @tufalabs soon.
- New SoTA on ARC-AGI. Nothing like the synergy of an awesome team (@bayesilicon @MohamedOsmanML). From 53 to 54.5 today. Onward and upward! 🚀 @arcprize @mikeknoop @fchollet @bryanlanders @GregKamradt @MLStreetTalk First. #kaggle -
- Excited to advance our lead and SoTA score on ARC-AGI-2 (@arcprize) by 3 points to 15.28. @DriesSmit1 @MohamedOsmanML @bayesilicon @GregKamradt @tufalabs
- ARC-AGI update. We posted a 41% on the official leaderboard! Thanks to my great team! @bayesilicon @epsil0ndelta Here's to hoping and working towards an even higher score. 🙏
- This is very interesting work. But if we are going to play the SoTA game, our solution that first broke SoTA on the private test set (34%) for ARC scored 60% on the public test set back in February, 2024. You wouldn't want us to run it on the training dataset like Ryan did.ARC-AGI’s been hyped over the last week as a benchmark that LLMs can’t solve. This claim triggered my dear coworker Ryan Greenblatt so he spent the last week trying to solve it with LLMs. Ryan gets 71% accuracy on a set of examples where humans get 85%; this is SOTA.
- 49.5 on ARC-AGI (gained .5 points). Is that 50% level magical, or will it fall soon? Our fantastic @bayesilicon is cooking something hot, and our superb @MohamedOsmanML has some very cool stuff in the works. What a great team! @arcprize
- Can you feel the 50%? 🚀New ARC-AGI high score! 49% (Prize goal: 85%) Congratulations, MindsAI!
- Always love to hear @fchollet. He has created many years of hard challenges for people and machines. More importantly, it is advancing the science of AI. ARC-AGI is one of his brilliant creations and a steppingstone towards AGI.
00:00 - Replying to @neurosp1keI found some time ago that it was good to think of the interview being reversed. In other words, what does the choice of interview process say about the environment of the organization. How did I feel as a result the interaction? Would I want to work with people who treat
- Excited to be full time on ARC-AGI-2. It's going to be a slow cook for the new ideas to be realized. Maybe we'll see a new face or two on the team! 🚀 @arcprize @MohamedOsmanML @bayesilicon @DriesSmit1
- 46% -- let's go!Woke up to find only a casual 3% improvement on the SOTA @arcprize today!! Huge kudos to our dream team @MindsAI_Jack @bayesilicon We're just getting started @GregKamradt @fchollet @mikeknoop
