(meaning-machine.streamlit.app)

114 points jdspiral | 1 comments | 22 Apr 25 22:55 UTC | HN request time: 0.342s | source

Show context

wrs ◴[23 Apr 25 03:48 UTC] No.43768465[source]▶

Is there evidence that modern LLMs identify parts of speech in an observable way? This explanation sounds more like how we did it in the 90s before deep learning took over.

replies(1): >>43769101 #

1. Xmd5a ◴[23 Apr 25 06:05 UTC] No.43769101[source]▶

>>43768465 #

https://arxiv.org/abs/1906.04341

https://arxiv.org/abs/1905.05950

https://en.wikiversity.org/wiki/Psycholinguistics/Models_of_...

↑

Meaning Machine – Visualize how LLMs break down and simulate meaning