tried o1-preview on @arcprize
result: 1 out of 2 tests correct
so o1-preview isn't going to solve 100% ARC Prize tasks
tbd on what % it gets compared to SOTA approaches, still testing rest of the lot
We put OpenAI o1 to the test against ARC Prize.
Results: both o1 models beat GPT-4o. And o1-preview is on par with Claude 3.5 Sonnet.
Can chain-of-thought scale to AGI? What explains o1's modest scores on ARC-AGI?
Our notes:
https://t.co/sV6LM1foGx
@GoGamingNL hey I’m trying to get customer support regarding a paid booking tomorrow, but you’re not responding (email on Thursday). Can you please reach out?
@TelenetEN It’s fixed now. It was mainly a comment to your approach of first line support only checking modem, and sending customer packing if modem works fine. I don’t seem to be getting anywhere with this, so nevermind. Thanks anyway.
@TelenetEN hey reached out this morning to technicall support by phone reporting VPN connectivity issues which were affecting me and other work colleagues. Support person insisted all was ok with my connection (mode check) then hung up on me.
@TelenetEN Hey, the support person had same reflex. In meantime my company IT opened a business ticket with Telenet, many customers complained. Telenet then fixed issue. How can you make sure to escalate tickets to technical teams so you notice your error instead of blocking it frontline ?
Throwback to 5 years ago when Tony Fauci, at 74 yo, was suiting up to treat an Ebola patient himself because he "wanted to show his staff that he wouldn't ask them to do anything he wouldn't do himself". This is what leadership looks like. https://t.co/QctW672ykC
@HelenRyles Hey do you know about the concept of multipotentialite?
Lots of projects, without going for one thing in depth, focus on persuing new interest.
Watch this TED talk: https://t.co/RYW2EIeYIf