Learning Theory

Meta Optimality for Demographic Parity Constrained Regression via Post-Processing ICML 2025 Metric Learning in an RKHS UAI 2025 MGDA Converges under Generalized Smoothness, Provably ICLR 2025 Mind the Gap: A Closer Look at Tokenization for Multiple-Choice Question Answering with LLMs EMNLP 2025 Mind the Gap: How BabyLMs Learn Filler-Gap Dependencies EMNLP 2025 Minimax-optimal and Locally-adaptive Online Nonparametric Regression ALT 2025 Minimum Empirical Divergence for Sub-Gaussian Linear Bandits AISTATS 2025 Minimum Width for Universal Approximation using Squashable Activation Functions ICML 2025 Misspecified Q-Learning with Sparse Linear Function Approximation: Tight Bounds on Approximation Error ICLR 2025 Mixture of Experts Provably Detect and Learn the Latent Cluster Structure in Gradient-Based Learning ICML 2025 MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark ACL 2025 Model selection for behavioral learning data and applications to contextual bandits AISTATS 2025 Model Uncertainty Quantification by Conformal Prediction in Continual Learning ICML 2025 Multi-agent Multi-armed Bandit Regret Complexity and Optimality AISTATS 2025 Multi-Pass Memory Lower Bounds for Learning Problems COLT 2025 Narrowing the Gap between Adversarial and Stochastic MDPs via Policy Optimization AISTATS 2025 Nearly Optimal Algorithms for Contextual Dueling Bandits from Adversarial Feedback ICML 2025 Nearly Optimal Differentially Private ReLU Regression UAI 2025 Nearly Optimal Sample Complexity for Learning with Label Proportions ICML 2025 Nearly Tight Bounds for Exploration in Streaming Multi-Armed Bandits with Known Optimality Gap AAAI 2025 NEAR: Neural Electromagnetic Array Response ICML 2025 Near-optimal Active Regression of Single-Index Models ICLR 2025 NesTools: A Dataset for Evaluating Nested Tool Learning Abilities of Large Language Models COLING 2025 Neural Collapse Beyond the Unconstrained Features Model: Landscape, Dynamics, and Generalization in the Mean-Field Regime ICML 2025 Neural Networks Perform Sufficient Dimension Reduction AAAI 2025