CLIP — our new neural network for classifying images into categories based on arbitrary text. Works like GPT-3's semantic search, but for images: andrewmayneblog.wordpress.co… Maybe a path towards vision nets that don't make silly mistakes: openai.com/blog/clip/

8:52 PM · Jan 5, 2021

15
103
10
488
Replying to @gdb
Seriously awesome.
0
0
0
0
Replying to @gdb
awesome
0
0
0
0
Replying to @gdb
Amazing work Greg. I am looking forward to see the results of what GTP3 can bring to the world.
0
0
0
0
Replying to @gdb
Pretty interesting, awesome work @gdb 👍
0
0
0
0
Replying to @gdb
Great work - amazing to see transformers being applied to image analysis. This could push the state of the art much further down the long tail.
0
0
0
1
Replying to @gdb
If Ai not only learns the meaning behind words but also people’s intentions behind those words, it will surpass current human communication
0
0
0
0
Replying to @gdb
Hi Greg, we have applied for GPT-3 many times during the last few months but haven't had success. I think we have a very legitimate use-case that can impact productivity for many people. Can we get access? :)
0
0
0
0
Replying to @gdb
Can we try?
0
0
0
0