4 ms·
Models have been trained on synthetic inputs for over a year now. Collapse only happens in very specific circumstances, it's not a limit in practice.
by int_19h 15d ago
Models have been trained on synthetic inputs for over a year now. Collapse only happens in very specific circumstances, it's not a limit in practice.
- johnsmith1840 14d agoYeah, that's why fully autonomous robots are so easy to train on novel tasks. Or why figure is strapping cameras to human workers for cleaning. Sythetic data works in a bounded box we have mapped you can't just make up novel data and it works. Model collapse is completely unsolved.