Machine Learning
Papers on all aspects of machine learning research (supervised, unsupervised, reinforcement learning, bandit problems, and so on) including also robustness, explanation, fairness, and methodology. cs.LG is also an appropriate primary category for applications of machine learning methods.
ROME and MEMIT produce fluent multi-hop answers when evaluated with chat templates, revealing that the field's consensus on poor…
· “Generalizable Model Editing Could Be More Than Direct Generalization”
By using the polar decomposition, the method produces exact p-values for differences between clusters under any unknown data covariance.
· “Selective Inference for Clustering Under Unknown Covariance”
Two heads exactly denoise a Gaussian mixture; linear attention cannot approach the Bayes risk regardless of heads.
· “Softmax Attention on Gaussian Mixtures: Linear When It Can, Selective When It Must”
