Re:Latent Dictionary: 3D map of Oxford3000+search words via ...
by pyinstallwoes on Saturday December 30, 2023 @03:31PM
(#38818807)
Yeah but it doesn’t use a universal method does it? And it requires labeling.
The method I’m describing requires no labeling. Labeling would be a local only translation (alias). Labels emerge based on meaning. But the labels are more of an interface - not the actual nodes themselves which arise off the not identity principle * event proximity * comparisons.
Re:Latent Dictionary: 3D map of Oxford3000+search words via ...
by warkdarrior on Saturday December 30, 2023 @10:52PM
(#38821595)
Labeling (which is typically manual and thus not scalable) is a proxy for comparisons. Two things are the same if they have the same label. The question is how else to encode the comparison information.
Re:Latent Dictionary: 3D map of Oxford3000+search words via ...
by pyinstallwoes on Thursday January 04, 2024 @04:29AM
(#38864975)
Right one is manual the other is automatic and my hypothesis is you can have automatic universal labeling the way I describe
That is how CLIP embeddings work and were trained to work.
Hugging Face transformers now has a get_image_features() and get_text_features() function for CLIP models to make getting the embeddings for different modalities easy: https://huggingface.co/docs/transformers/model_doc/clip#tran...