4 ms·
This is the definition of training the model on it's own output. Apparently that is all ok now.
by consp 1y ago
This is the definition of training the model on it's own output. Apparently that is all ok now.
- baby 1y agoI mean you're supposed to use RAG to avoid hallucinations
- MagicMoonlight 1y agoYeah they call it “synthetic data” and wonder why their models are slop now