Learning Theory

Principled Understanding of Generalization for Generative Transformer Models in Arithmetic Reasoning Tasks ACL 2025 Privacy Amplification Through Synthetic Data: Insights from Linear Regression ICML 2025 Private List Learnability vs. Online List Learnability COLT 2025 Private Realizable-to-Agnostic Transformation with Near-Optimal Sample Complexity COLT 2025 PRMBench: A Fine-grained and Challenging Benchmark for Process-Level Reward Models ACL 2025 Probing Neural Network Generalization using Default Patterns NAACL 2025 Probing the Mid-level Vision Capabilities of Self-Supervised Learning CVPR 2025 Problem-dependent Regret for Lexicographic Multi-Armed Bandits with Adversarial Corruptions IJCAI 2025 Procedural Knowledge in Pretraining Drives Reasoning in Large Language Models ICLR 2025 Provable Length Generalization in Sequence Prediction via Spectral Filtering ICML 2025 Provable Maximum Entropy Manifold Exploration via Diffusion Models ICML 2025 Provable unlearning in topic modeling and downstream tasks ICLR 2025 Provable weak-to-strong generalization via benign overfitting ICLR 2025 Provably Adaptive Average Reward Reinforcement Learning for Metric Spaces UAI 2025 Provably Cost-Sensitive Adversarial Defense via Randomized Smoothing ICML 2025 Provably Efficient Exploration in Inverse Constrained Reinforcement Learning ICML 2025 Provably Efficient RL for Linear MDPs under Instantaneous Safety Constraints in Non-Convex Feature Spaces ICML 2025 Provably Improving Generalization of Few-shot models with Synthetic Data ICML 2025 Provably Near-Optimal Federated Ensemble Distillation with Negligible Overhead ICML 2025 QPRL : Learning Optimal Policies with Quasi-Potential Functions for Asymmetric Traversal ICML 2025 Quantification of Large Language Model Distillation ACL 2025 Quantifying Generalization Complexity for Large Language Models ICLR 2025 Quantifying Overfitting along the Regularization Path for Two-Part-Code MDL in Supervised Classification COLT 2025 Quantile Multi-Armed Bandits with 1-bit Feedback ALT 2025 Quantum-inspired Non-homologous Representation Constraint Mechanism for Long-tail Senses of Word Sense Disambiguation AAAI 2025