Learning Theory

CAPTURE: Evaluating Spatial Reasoning in Vision Language Models via Occluded Object Counting ICCV 2025 Capturing the Temporal Dependence of Training Data Influence ICLR 2025 CARL: Causality-guided Architecture Representation Learning for an Interpretable Performance Predictor ICCV 2025 Cascaded Information Disclosure for Generalized Evaluation of Problem Solving Capabilities IJCNLP 2025 CAST: Cross-modal Alignment Similarity Test for Vision Language Models COLING 2025 CDT: A Comprehensive Capability Framework for Large Language Models Across Cognition, Domain, and Task EMNLP 2025 Certifiably Robust Model Evaluation in Federated Learning under Meta-Distributional Shifts ICML 2025 Certification of Speaker Recognition Models to Additive Perturbations AAAI 2025 Certified Unlearning for Neural Networks ICML 2025 Child-Directed Language Does Not Consistently Boost Syntax Learning in Language Models EMNLP 2025 Circuit Complexity Bounds for RoPE-based Transformer Architecture EMNLP 2025 Circuit Compositions: Exploring Modular Structures in Transformer-Based Language Models ACL 2025 Circuit Stability Characterizes Language Model Generalization ACL 2025 CLDyB: Towards Dynamic Benchmarking for Continual Learning with Pre-trained Models ICLR 2025 Clipping Improves Adam-Norm and AdaGrad-Norm when the Noise Is Heavy-Tailed ICML 2025 Compelling ReLU Networks to Exhibit Exponentially Many Linear Regions at Initialization and During Training ICML 2025 Competing Bandits in Matching Markets via Super Stability ICML 2025 Competition Dynamics Shape Algorithmic Phases of In-Context Learning ICLR 2025 Complexity Lower Bounds of Adaptive Gradient Algorithms for Non-convex Stochastic Optimization under Relaxed Smoothness ICLR 2025 Complexity of Injectivity and Verification of ReLU Neural Networks (Extended Abstract) COLT 2025 Computable learning of natural hypothesis classes COLT 2025 Computationally Efficient RL under Linear Bellman Completeness for Deterministic Dynamics ICLR 2025 Compute-Optimal LLMs Provably Generalize Better with Scale ICLR 2025 Compute Optimal Scaling of Skills: Knowledge vs Reasoning ACL 2025 Computing High-dimensional Confidence Sets for Arbitrary Distributions COLT 2025