Content
@
https://warpcast.com/~/channel/aichannel
0 reply
0 recast
0 reaction
shoni.eth
@alexpaden
this creates an interesting dilemma: a ‘large language model’ is a subset of ‘AI’, but if you abstract too deeply, you risk losing what’s novel about a trending topic—‘super bowl sunday’ becomes just ‘football.’ as you capture more content, you lose nuance; as you capture more nuance, you find less content—at least on the first pass. how would you approach this problem, assuming a small LLM is generating the classification in just a few words?
3 replies
0 recast
0 reaction
Dan Romero
@dwr.eth
Human editors
0 reply
0 recast
1 reaction