Similarity Learning lacks a framework. So we built one

Re:Similarity Learning lacks a framework. So we built one by dmos62 on Wednesday July 13, 2022 @10:45AM (#32083469)

I realise this is an overly-broad question, but any insight into what's the state-of-art in Similarity Learning for article-type text?

More specifically, I'm interested in deriving distances between writing style, arguing style, etc.

[ Parent | on HN ]

Re:Similarity Learning lacks a framework. So we built one by monatis on Wednesday July 13, 2022 @11:20AM (#32083921)

There's study here: http://cs230.stanford.edu/projects_spring_2021/reports/57.pd...

Basically, you can collect text from different authors, then you can use authors names as labels to train a similarity learning with it. My suggestion would be finetune a Transformer model with a specific head and an ArcFace loss.

[ Parent | on HN ]

Re:Similarity Learning lacks a framework. So we built one by generall on Wednesday July 13, 2022 @10:58AM (#32083617)

It is definitely possible to do, if you have a proper training set. You would need to somehow give model a signal, that you are interesting in e.g. arguing style specifically, and not the topic of the text.

[ Parent | on HN ]