Simons Institute Video Watch posted an update
What Kind of Hallucination Is This? Decomposing Uncertainty for Generative Models
Why it mattersThe talk presents a framework for estimating uncertainty in generative models, including how kernel scores can derive uncertainty measures from a finite set of outputs. It also describes separating ambiguity in a prompt from a model's lack of knowledge, then distilling those estimates into calibrated uncertainty.
Discuss: How might separating prompt ambiguity from a model's lack of knowledge change how uncertainty is communicated in clinical decision support?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.
Your turn
Pull up a chair.
Write first. We’ll sort the introductions when you submit.