Simons Institute Video Watch posted an update
Quantifying Reliability in AI: Hallucinations and Uncertainty
Why it mattersThe talk frames quantifying reliability as central to trustworthy AI, then describes feature-based conformal techniques for detecting and quantifying hallucinations in image reconstruction and probabilistic regression with diffusion models for uncertainty quantification. It also raises open mathematical questions about when model outputs can be trusted.
Discuss: How might feature-based conformal techniques for image-reconstruction hallucinations and diffusion-model uncertainty quantification change when an AI output is considered reliable?
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.