#classification

2 stories

#technology #calibration #email #feature engineering #irony detection #jev #large language models #local #logistic regression #machine learning #model interpretability #nobodywho

Jev in 25 lines of Python
A 25-line Python model that classifies emails as legitimate, spam, or phishing without sending data.
NobodyWho · 1m
Stop using LLMs as classifiers
LLMs seem good at classification but lack calibration, interpretability, and control. Treat them as feature engineers instead.
Minimally Sufficient · 3m