News for hackers. Stuff that matters.
Re:An analysis of DeepSeek's R1-Zero and R1 by optimalsolver on Wednesday January 29, 2025 @01:42PM (#42869265)
But can o3 write a symphony?

Seriously though, I'd like to hear suggestions on how to automatically evaluate an AI model's creativity, no humans in the loop.

Re:An analysis of DeepSeek's R1-Zero and R1 by baq on Wednesday January 29, 2025 @03:44PM (#42870918)
LLMs have read everything humans made so just ask one if there’s anything truly new in that freshly confabulated slop-phony.
Re:An analysis of DeepSeek's R1-Zero and R1 by johnfn on Wednesday January 29, 2025 @02:35PM (#42870011)
Have you tried suno.ai?
Re:An analysis of DeepSeek's R1-Zero and R1 by Vampiero on Wednesday January 29, 2025 @07:05PM (#42873089)
Have _you_? It lost its novelty after a couple of days.
Re:An analysis of DeepSeek's R1-Zero and R1 by drusepth on Thursday January 30, 2025 @05:53PM (#42883035)
I probably listen to Suno (both my own songs, and songs other people have created) about as often as I listen to Spotify, these days.
Re:An analysis of DeepSeek's R1-Zero and R1 by fragmede on Wednesday January 29, 2025 @01:59PM (#42869487)
we'd have to create a numerical scale for creativity, from boring to Dali, with milliEschers and MegaGeigers somewhere in there as well
Re:An analysis of DeepSeek's R1-Zero and R1 by rpastuszak on Wednesday January 29, 2025 @02:23PM (#42869839)
It's essential that we quantify everything so that we can put a price on it. I'd go with Kahlograms though.
Re:An analysis of DeepSeek's R1-Zero and R1 by gsam on Wednesday January 29, 2025 @04:31PM (#42871446)
In my view there's two modes of creativity:

1. That two distant topics or ideas are actually much more closely related. The creative sees one example of an idea and applies it to a discipline that nobody expects. In theory, reduction of the maximally distant can probably be measured with a tangible metric.

2. Discovery of ideas that are even more maximally distant. Pushing the edge, and this can be done by pure search and randomness actually. But it's no good if it's garbage. The trick is, what is garbage? That is very context dependent.

(Also, a creative might be measured on the efficiency of these metrics rather than absolute output)

Re:An analysis of DeepSeek's R1-Zero and R1 by docfort on Thursday January 30, 2025 @07:20AM (#42877087)
Terry Tao has referred to this classification system as foxes vs hedgehogs. https://en.m.wikipedia.org/wiki/The_Hedgehog_and_the_Fox