Vision language models are blind

Re:Vision language models are blind by scarface_74 on Wednesday July 10, 2024 @07:38PM (#40932537)

Out of curiosity, I tried your test with ChatGPT 4o

Of course as usual, LLMs are horrible with Math.

Funny enough, the next time it verified the word count by counting it out until I specifically told it to use Python

[ Parent | on HN ]

Re:Vision language models are blind by infiar on Wednesday July 10, 2024 @09:43PM (#40933093)

This counting words task reminded me of a youtube video: https://www.youtube.com/watch?v=-9XKiOXaHlI Maybe LLMs are somehow more like monkeys.

[ Parent | on HN ]