Glyph: Scaling Context Windows via Visual-Text Compression(github.com)
github.com
Glyph: Scaling Context Windows via Visual-Text Compression
https://github.com/thu-coai/Glyph
3 comments
Can someone compare/contrast with deepseek-ocr?
This looks very promising. Are there any downsides or potential gotchas?
I asked this question on another post and was downvoted, trying again: don't we lose the "contextualization" that LLM embeddings do (embedding on Token X contains not just information about X, but also of all tokens that came before X in the context, causing different embedding for "flies" in "time flies like an arrow" vs "fruit flies like a banana")?
The image embeddings, as I currently understand, are just pixel values of a block of pixels.
What am I missing?
The image embeddings, as I currently understand, are just pixel values of a block of pixels.
What am I missing?