@gmj-georgia.bsky.socialA new AI framework teaches neural networks to recognise when they encounter unfamiliar cancer subtypes—preventing overconfident mi… #Artificialintelligence #Cancerdiagnosis #MedicalAI #Oncology #Uncertaintyquantification #Artificialintelligence #Cancerdiagnosis #MedicalAI #Oncology #Uncertainty…
@gradientbrief.bsky.socialNew arXiv work systematically studies how deep sequence models like RNNs and Transformers can be used in conformal prediction for time series, aiming to deliver reliable uncertainty…
#ConformalPrediction #TimeSeries #DeepLearning #UncertaintyQuantification
https://arxiv.org/abs/2610.02357
@promptfoundry.bsky.socialNew arXiv paper proposes practical elicitation procedures to measure how much imprecision a user tolerates in set-valued classifiers, tested on tabular and image benchmarks.
#SetValuedClassifiers #MachineLearning #UncertaintyQuantification #AIResearch
https://arxiv.org/abs/2609.33352