Learning Theory

RAGGED: Towards Informed Design of Scalable and Stable RAG Systems ICML 2025 Random Pruning Over-parameterized Neural Networks Can Improve Generalization: A Training Dynamics Analysis JMLR 2025 Rate of Model Collapse in Recursive Training AISTATS 2025 Rates for Offline Reinforcement Learning with Adaptively Collected Data L4DC 2025 Realistic Evaluation of Deep Partial-Label Learning Algorithms ICLR 2025 Reasoning Models Reason Well, Until They Don’t IJCNLP 2025 ReasoningWeekly: A General Knowledge and Verbal Reasoning Challenge for Large Language Models IJCNLP 2025 Reason to Rote: Rethinking Memorization in Reasoning EMNLP 2025 Recitation over Reasoning: How Cutting-Edge Language Models Can Fail on Elementary School-Level Reasoning Problems? IJCNLP 2025 Recite, Reconstruct, Recollect: Memorization in LMs as a Multifaceted Phenomenon ICLR 2025 Recursive Training Loops in LLMs: How training data properties modulate distribution shift in generated data? EMNLP 2025 Refined generalization analysis of the Deep Ritz Method and Physics-Informed Neural Networks ICML 2025 Refining the Sample Complexity of Comparative Learning ALT 2025 Regret Analysis of Multi-task Representation Learning for Linear-Quadratic Adaptive Control AAAI 2025 Regret-Free Reinforcement Learning for Temporal Logic Specifications ICML 2025 Regret-Optimal List Replicable Bandit Learning: Matching Upper and Lower Bounds ICLR 2025 Regularity in Canonicalized Models: A Theoretical Perspective AISTATS 2025 Regular-pattern-sensitive CRFs for Distant Label Interactions ACL 2025 Reinforcement Learning for Infinite-Horizon Average-Reward Linear MDPs via Approximation by Discounted-Reward MDPs AISTATS 2025 Rejecting Hallucinated State Targets during Planning ICML 2025 Relational Invariant Learning for Robust Solvation Free Energy Prediction ICML 2025 Reminiscence Attack on Residuals: Exploiting Approximate Machine Unlearning for Privacy ICCV 2025 Representations Shape Weak-to-Strong Generalization: Theoretical Insights and Empirical Predictions ICML 2025 Representative Language Generation ICML 2025 ResKoopNet: Learning Koopman Representations for Complex Dynamics with Spectral Residuals ICML 2025