{"total":91,"rows":[{"id":"2304.11158","title":"Emergent and Predictable Memorization in Large Language Models","created":"2023-04-21","cited_by_count":148,"area":null},{"id":"2503.02113","title":"Deep Learning is Not So Mysterious or Different","created":"2025-03-03","cited_by_count":56,"area":null},{"id":"2208.02813","title":"Towards Understanding Mixture of Experts in Deep Learning","created":"2022-08-04","cited_by_count":52,"area":"moe"},{"id":"2306.03435","title":"On the Role of Attention in Prompt-tuning","created":"2023-06-06","cited_by_count":50,"area":null},{"id":"2307.08286","title":"Going Beyond Linear Mode Connectivity: The Layerwise Linear Feature Connectivity","created":"2023-07-17","cited_by_count":50,"area":null},{"id":"2210.07082","title":"Implicit Bias in Leaky ReLU Networks Trained on High-Dimensional Data","created":"2022-10-13","cited_by_count":46,"area":null},{"id":"2306.04949","title":"Robust Learning with Progressive Data Expansion Against Spurious Correlation","created":"2023-06-08","cited_by_count":39,"area":null},{"id":"2504.10957","title":"When is Task Vector Provably Effective for Model Editing? A Generalization Analysis of Nonlinear Transformers","created":"2025-04-15","cited_by_count":38,"area":"safety"},{"id":"2207.06569","title":"Benign, Tempered, or Catastrophic: A Taxonomy of Overfitting","created":"2022-07-14","cited_by_count":36,"area":null},{"id":"2304.11327","title":"Understanding and Improving Feature Learning for Out-of-Distribution Generalization","created":"2023-04-22","cited_by_count":35,"area":null},{"id":"2303.04145","title":"Benign Overfitting for Two-layer ReLU Convolutional Neural Networks","created":"2023-03-07","cited_by_count":34,"area":null},{"id":"2310.02541","title":"Benign Overfitting and Grokking in ReLU Networks for XOR Cluster Data","created":"2023-10-04","cited_by_count":33,"area":null},{"id":"2402.05738","title":"Implicit Bias and Fast Convergence Rates for Self-attention","created":"2024-02-08","cited_by_count":31,"area":null},{"id":"2303.01462","title":"Benign Overfitting in Linear Classifiers and Leaky ReLU Networks from KKT Conditions for Margin Maximization","created":"2023-03-02","cited_by_count":29,"area":null},{"id":"2310.07269","title":"Why Does Sharpness-Aware Minimization Generalize Better Than SGD?","created":"2023-10-11","cited_by_count":28,"area":null},{"id":"2303.01456","title":"The Double-Edged Sword of Implicit Bias: Generalization vs. Robustness in ReLU Networks","created":"2023-03-02","cited_by_count":23,"area":null},{"id":"2409.19610","title":"Federated Learning from Vision-Language Foundation Models: Theoretical Analysis and Method","created":"2024-09-29","cited_by_count":22,"area":"mllm"},{"id":"2302.09235","title":"Generalization and Stability of Interpolating Neural Networks with Minimal Width","created":"2023-02-18","cited_by_count":21,"area":null},{"id":"2303.08433","title":"The Benefits of Mixup for Feature Learning","created":"2023-03-15","cited_by_count":19,"area":null},{"id":"2310.18935","title":"Implicit Bias of Gradient Descent for Two-layer ReLU and Leaky ReLU Networks on Nearly-orthogonal Data","created":"2023-10-29","cited_by_count":18,"area":null},{"id":"2305.15141","title":"From Tempered to Benign Overfitting in ReLU Neural Networks","created":"2023-05-24","cited_by_count":16,"area":null},{"id":"2406.03880","title":"Memorization in deep learning: A survey","created":"2024-06-06","cited_by_count":15,"area":null},{"id":"2310.10767","title":"Wide Neural Networks as Gaussian Processes: Lessons from Deep Equilibrium Models","created":"2023-10-16","cited_by_count":14,"area":null},{"id":"2310.12408","title":"Provable Guarantees for Neural Networks via Gradient Feature Learning","created":"2023-10-19","cited_by_count":14,"area":null},{"id":"2407.13621","title":"Differential Privacy Mechanisms in Neural Tangent Kernel Regression","created":"2024-07-18","cited_by_count":14,"area":null},{"id":"2412.02366","title":"GenMix: Effective Data Augmentation with Generative Diffusion Model Image Editing","created":"2024-12-03","cited_by_count":14,"area":"diffusion"},{"id":"2306.13926","title":"Quantifying the Optimization and Generalization Advantages of Graph Neural Networks Over Multilayer Perceptrons","created":"2023-06-24","cited_by_count":12,"area":null},{"id":"2206.13482","title":"Understanding Benign Overfitting in Gradient-Based Meta Learning","created":"2022-06-27","cited_by_count":11,"area":null},{"id":"2307.15396","title":"Noisy Interpolation Learning with Shallow Univariate ReLU Networks","created":"2023-07-28","cited_by_count":11,"area":null},{"id":"2309.04268","title":"Optimal Rate of Kernel Regression in Large Dimensions","created":"2023-09-08","cited_by_count":11,"area":null},{"id":"2408.16939","title":"Theoretical Insights into Overparameterized Models in Multi-Task and Replay-Based Continual Learning","created":"2024-08-29","cited_by_count":11,"area":null},{"id":"2602.16340","title":"The Implicit Bias of Adam and Muon on Smooth Homogeneous Neural Networks","created":"2026-02-18","cited_by_count":11,"area":null},{"id":"2305.19377","title":"Benign Overfitting in Deep Neural Networks under Lazy Training","created":"2023-05-30","cited_by_count":10,"area":null},{"id":"2310.01975","title":"Benign Overfitting in Two-Layer ReLU Convolutional Neural Networks for XOR Data","created":"2023-10-03","cited_by_count":10,"area":null},{"id":"2206.09136","title":"Provable Generalization of Overparameterized Meta-learning Trained with SGD","created":"2022-06-18","cited_by_count":9,"area":null},{"id":"2210.14184","title":"Learning Ability of Interpolating Deep Convolutional Neural Networks","created":"2022-10-25","cited_by_count":9,"area":null},{"id":"2412.01256","title":"NLPrompt: Noise-Label Prompt Learning for Vision-Language Models","created":"2024-12-02","cited_by_count":9,"area":"mllm"},{"id":"2209.09315","title":"Deep Linear Networks can Benignly Overfit when Shallow Ones Do","created":"2022-09-19","cited_by_count":8,"area":null},{"id":"2305.09947","title":"Understanding the Initial Condensation of Convolutional Neural Networks","created":"2023-05-17","cited_by_count":8,"area":null},{"id":"2306.11680","title":"The Implicit Bias of Batch Normalization in Linear Models and Two-layer Linear Convolutional Neural Networks","created":"2023-06-20","cited_by_count":8,"area":null},{"id":"2207.11621","title":"A Universal Trade-off Between the Model Size, Test Loss, and Training Loss of Linear Predictors","created":"2022-07-23","cited_by_count":7,"area":null},{"id":"2303.17940","title":"Per-Example Gradient Regularization Improves Learning Signals from Noisy Data","created":"2023-03-31","cited_by_count":7,"area":null},{"id":"2306.09955","title":"Training shallow ReLU networks on noisy data using hinge loss: when do we overfit and is it benign?","created":"2023-06-16","cited_by_count":7,"area":null},{"id":"2403.17592","title":"On the Benefits of Over-parameterization for Out-of-Distribution Generalization","created":"2024-03-26","cited_by_count":7,"area":null},{"id":"2409.03891","title":"Overfitting Behaviour of Gaussian Kernel Ridgeless Regression: Varying Bandwidth or Dimensionality","created":"2024-09-05","cited_by_count":7,"area":null},{"id":"2410.23495","title":"DASH: Warm-Starting Neural Network Training in Stationary Settings without Loss of Plasticity","created":"2024-10-30","cited_by_count":7,"area":null},{"id":"2206.07892","title":"Max-Margin Works while Large Margin Fails: Generalization without Uniform Convergence","created":"2022-06-16","cited_by_count":6,"area":null},{"id":"2403.06903","title":"Benign overfitting in leaky ReLU networks with moderate input dimension","created":"2024-03-11","cited_by_count":6,"area":null},{"id":"2206.00501","title":"Benign Overfitting in Classification: Provably Counter Label Noise with Larger Models","created":"2022-06-01","cited_by_count":5,"area":null},{"id":"2406.03944","title":"Provably Neural Active Learning Succeeds via Prioritizing Perplexing Samples","created":"2024-06-06","cited_by_count":5,"area":null},{"id":"2407.05594","title":"SLIM: Spuriousness Mitigation with Minimal Human Annotations","created":"2024-07-08","cited_by_count":5,"area":null},{"id":"2410.19092","title":"Provable Tempered Overfitting of Minimal Nets and Typical Nets","created":"2024-10-24","cited_by_count":5,"area":null},{"id":"2502.19921","title":"Shifting the Paradigm: A Diffeomorphism Between Time Series Data Manifolds for Achieving Shift-Invariancy in Deep Learning","created":"2025-02-27","cited_by_count":5,"area":null},{"id":"2410.23672","title":"Provable Benefit of Cutout and CutMix for Feature Learning","created":"2024-10-31","cited_by_count":4,"area":null},{"id":"2505.18909","title":"On the Role of Label Noise in the Feature Learning Process","created":"2025-05-25","cited_by_count":4,"area":null},{"id":"2510.17526","title":"How Does Label Noise Gradient Descent Improve Generalization in the Low SNR Regime?","created":"2025-10-20","cited_by_count":4,"area":null},{"id":"2206.05997","title":"Analysis of function approximation and stability of general DNNs in directed acyclic graphs using un-rectifying analysis","created":"2022-06-13","cited_by_count":3,"area":null},{"id":"2208.08003","title":"Investigating the Impact of Model Width and Density on Generalization in Presence of Label Noise","created":"2022-08-17","cited_by_count":3,"area":null},{"id":"2301.00335","title":"Pruning Before Training May Improve Generalization, Provably","created":"2023-01-01","cited_by_count":3,"area":null},{"id":"2303.16887","title":"Towards Understanding the Effect of Pretraining Label Granularity","created":"2023-03-29","cited_by_count":3,"area":null},{"id":"2404.17768","title":"Changing the Training Data Distribution to Reduce Simplicity Bias Improves In-distribution Generalization","created":"2024-04-27","cited_by_count":3,"area":null},{"id":"2501.17151","title":"Scanning Trojaned Models Using Out-of-Distribution Samples","created":"2025-01-28","cited_by_count":3,"area":"safety"},{"id":"2504.09347","title":"Inference for Deep Neural Network Estimators in Generalized Nonparametric Models","created":"2025-04-12","cited_by_count":3,"area":null},{"id":"2511.18006","title":"Understanding Private Learning From Feature Perspective","created":"2025-11-22","cited_by_count":3,"area":null},{"id":"2406.01435","title":"Learning Analysis of Kernel Ridgeless Regression with Asymmetric Kernel Learning","created":"2024-06-03","cited_by_count":2,"area":null},{"id":"2409.18685","title":"Understanding the Benefits of SimCLR Pre-Training in Two-Layer Convolutional Neural Networks","created":"2024-09-27","cited_by_count":2,"area":null},{"id":"2502.11893","title":"Rethinking Benign Overfitting in Two-Layer Neural Networks","created":"2025-02-17","cited_by_count":2,"area":null},{"id":"2502.12508","title":"Understanding Generalization in Transformers: Error Bounds and Training Dynamics Under Benign and Harmful Overfitting","created":"2025-02-18","cited_by_count":2,"area":null},{"id":"2505.21574","title":"Do We Need All the Synthetic Data? Targeted Image Augmentation via Diffusion Models","created":"2025-05-27","cited_by_count":2,"area":"diffusion"},{"id":"2509.18001","title":"Unveiling m-Sharpness Through the Structure of Stochastic Gradient Noise","created":"2025-09-22","cited_by_count":2,"area":null},{"id":"2606.23364","title":"Convergence of Gradient Descent for General Neural Network Architectures Beyond the NTK Regime","created":"2026-06-22","cited_by_count":2,"area":null},{"id":"2401.13624","title":"Can overfitted deep neural networks in adversarial training generalize? -- An approximation viewpoint","created":"2024-01-24","cited_by_count":1,"area":null},{"id":"2502.16076","title":"Harnessing Feature Resonance under Arbitrary Target Alignment for Out-of-Distribution Node Detection","created":"2025-02-22","cited_by_count":1,"area":null},{"id":"2506.24120","title":"Data Uniformity Improves Training Efficiency and More, with a Convergence Framework Beyond the NTK Regime","created":"2025-06-30","cited_by_count":1,"area":null},{"id":"2510.12026","title":"Mamba Can Learn Low-Dimensional Targets In-Context via Test-Time Feature Learning","created":"2025-10-14","cited_by_count":1,"area":"ssm"},{"id":"2602.00825","title":"Harmful Overfitting in Sobolev Spaces","created":"2026-01-31","cited_by_count":1,"area":null},{"id":"2602.02767","title":"Provable Effects of Data Replay in Continual Learning: A Feature Learning Perspective","created":"2026-02-02","cited_by_count":1,"area":null},{"id":"2602.03872","title":"Understanding the Impact of Differentially Private Training on Memorization of Long-Tailed Data","created":"2026-02-01","cited_by_count":1,"area":null},{"id":"2605.06314","title":"When Does $\\ell_2$-Boosting Overfit Benignly? High-Dimensional Risk Asymptotics and the $\\ell_1$ Implicit Bias","created":"2026-05-07","cited_by_count":1,"area":null},{"id":"2307.10736","title":"Long-Tail Theory under Gaussian Mixtures","created":"2023-07-20","cited_by_count":0,"area":null},{"id":"2404.04498","title":"Bayesian Inference for Consistent Predictions in Overparameterized Nonlinear Regression","created":"2024-04-06","cited_by_count":0,"area":null},{"id":"2409.00733","title":"Benign Overfitting under Learning Rate Conditions for $\\alpha$ Sub-exponential Input","created":"2024-09-01","cited_by_count":0,"area":null},{"id":"2504.08628","title":"Gradient Descent Robustly Learns the Intrinsic Dimension of Data in Training Convolutional Neural Networks","created":"2025-04-11","cited_by_count":0,"area":null},{"id":"2504.20617","title":"Sobolev norm inconsistency of kernel interpolation","created":"2025-04-29","cited_by_count":0,"area":null},{"id":"2505.16204","title":"Directional Convergence, Benign Overfitting of Gradient Descent in leaky ReLU two-layer Neural Networks","created":"2025-05-22","cited_by_count":0,"area":null},{"id":"2511.00480","title":"FedMGP: Personalized Federated Learning with Multi-Group Text-Visual Prompts","created":"2025-11-01","cited_by_count":0,"area":"mllm"},{"id":"2601.08316","title":"Deep Exploration of Epoch-wise Double Descent in Noisy Data: Signal Separation, Large Activation, and Benign Overfitting","created":"2026-01-13","cited_by_count":0,"area":null},{"id":"2607.09547","title":"High-Dimensional Interpolators Can Be Fragile: Heavy Tails and High-Dimensional Large Deviations","created":"2026-07-10","cited_by_count":0,"area":null},{"id":"2608.04382","title":"Non-asymptotic implicit bias of logistic regression at early-stage gradient descent dynamics","created":"2026-08-05","cited_by_count":0,"area":null},{"id":"2610.00403","title":"The Conflict Between Logic and Memory: Training Conditions for Optimizer-Dependent Rule Acquisition","created":"2026-09-30","cited_by_count":0,"area":null},{"id":"2610.00751","title":"Signal-Noise Factorization Isolates Nuisance Variation into Removable Subspaces","created":"2026-09-30","cited_by_count":0,"area":null}],"next_cursor":null}