Open Flamingo – open framework to train multimodal LLMs

Re:Open Flamingo – open framework to train multimodal LLMs by ftxbro on Tuesday March 28, 2023 @05:00PM (#35347768)

In the demo I put the obama prank photo http://karpathy.github.io/2012/10/22/state-of-computer-visio... and asked "Why is this picture funny?" and it responded "Question: Why is this picture funny? Answer: President Obama is taller than the average person."

[ Parent | on HN ]

Re:Open Flamingo – open framework to train multimodal LLMs by airstrike on Tuesday March 28, 2023 @05:31PM (#35348203)

What's the GPT-4 answer to why that picture is funny?

[ Parent | on HN ]

Re:Open Flamingo – open framework to train multimodal LLMs by vagabund on Tuesday March 28, 2023 @05:33PM (#35348232)

> @karpathy: We tried and it solves it :O. The vision capability is very strong but I still didn't believe it could be true. The waters are muddied some by a fear that my original post (or derivative work there of) is part of the training set. More on it later.

https://twitter.com/karpathy/status/1635697741925064704

[ Parent | on HN ]

Re:Open Flamingo – open framework to train multimodal LLMs by meghan_rain on Wednesday March 29, 2023 @12:23AM (#35352526)

Is there a More on it?

[ Parent | on HN ]

Re:Open Flamingo – open framework to train multimodal LLMs by codetrotter on Tuesday March 28, 2023 @05:51PM (#35348461)

Tbh I’m not sure why the pic is funny. And I am human

[ Parent | on HN ]

Re:Open Flamingo – open framework to train multimodal LLMs by david2ndaccount on Wednesday March 29, 2023 @02:05AM (#35353247)

That’s because it’s very posed so it’s not that funny.

[ Parent | on HN ]

Re:Open Flamingo – open framework to train multimodal LLMs by pretendscholar on Tuesday March 28, 2023 @06:17PM (#35348776)

Bot detected

[ Parent | on HN ]

Re:Open Flamingo – open framework to train multimodal LLMs by mylons on Tuesday March 28, 2023 @05:53PM (#35348489)

he's stepping on the scale making the guy seem heavier than he is

[ Parent | on HN ]

Re:Open Flamingo – open framework to train multimodal LLMs by pretendscholar on Tuesday March 28, 2023 @06:18PM (#35348788)

Furthermore the man on the scale is faced the other way and wouldn’t know someone is stepping on the scale. There’s an element of theory of mind there. You would have to understand that the man on the scale is unaware of Obama’s action.

[ Parent | on HN ]

Re:Open Flamingo – open framework to train multimodal LLMs by owlninja on Tuesday March 28, 2023 @10:47PM (#35351635)

The article points this out and several things we all instantly recognize.

[ Parent | on HN ]

Re:Open Flamingo – open framework to train multimodal LLMs by codetrotter on Wednesday March 29, 2023 @01:18AM (#35352967)

Thanks. I was looking at the picture on my phone and it was difficult to see what was going on in the picture.

[ Parent | on HN ]

Re:Open Flamingo – open framework to train multimodal LLMs by airstrike on Tuesday March 28, 2023 @05:53PM (#35348488)

Luckily the article explains it right below the caption ;-)

[ Parent | on HN ]

Re:Open Flamingo – open framework to train multimodal LLMs by elcomet on Wednesday March 29, 2023 @05:47AM (#35354803)

But you understand why people might find it funny, and what people in the image are thinking, you have a mental model of them. That's the point.

[ Parent | on HN ]

Re:Open Flamingo – open framework to train multimodal LLMs by totetsu on Wednesday March 29, 2023 @06:43AM (#35355236)

Is it something to do with the Idiom of putting a thumb on the scale, meaning using your power to influe the outcome of something?

[ Parent | on HN ]

Re:Open Flamingo – open framework to train multimodal LLMs by joenot443 on Wednesday March 29, 2023 @07:59AM (#35355874)

Has this one fared better with other image prompt models? It’s a great little challenge, I’m curious for a follow up!

[ Parent | on HN ]