{"total":58,"rows":[{"id":"2209.10652","title":"Toy Models of Superposition","created":"2022-09-21","cited_by_count":971,"area":"mechinterp"},{"id":"2106.09647","title":"Deep Learning Through the Lens of Example Difficulty","created":"2021-06-17","cited_by_count":141,"area":null},{"id":"2206.04041","title":"Neural Collapse: A Review on Modelling Principles and Generalization","created":"2022-06-08","cited_by_count":95,"area":null},{"id":"2406.11813","title":"How Do Large Language Models Acquire Factual Knowledge During Pretraining?","created":"2024-06-17","cited_by_count":90,"area":null},{"id":"2110.13905","title":"Gradient Descent on Two-layer Nets: Margin Maximization and Simplicity Bias","created":"2021-10-26","cited_by_count":83,"area":null},{"id":"2201.07395","title":"Overview frequency principle/spectral bias in deep learning","created":"2022-01-19","cited_by_count":69,"area":null},{"id":"2207.12678","title":"Analyzing Sharpness along GD Trajectory: Progressive Sharpening and Edge of Stability","created":"2022-07-26","cited_by_count":63,"area":null},{"id":"2211.08422","title":"Mechanistic Mode Connectivity","created":"2022-11-15","cited_by_count":50,"area":null},{"id":"2201.12498","title":"Investigating Why Contrastive Learning Benefits Robustness Against Label Noise","created":"2022-01-29","cited_by_count":46,"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":"2305.18761","title":"Identifying Spurious Biases Early in Training through the Lens of Simplicity Bias","created":"2023-05-30","cited_by_count":44,"area":null},{"id":"2310.09528","title":"Hypernetwork-based Meta-Learning for Low-Rank Physics-Informed Neural Networks","created":"2023-10-14","cited_by_count":44,"area":null},{"id":"2301.11500","title":"Understanding Incremental Learning of Gradient Descent: A Fine-grained Analysis of Matrix Sensing","created":"2023-01-27","cited_by_count":42,"area":null},{"id":"2205.10217","title":"Memorization and Optimization in Deep Neural Networks with Minimum Over-parameterization","created":"2022-05-20","cited_by_count":36,"area":null},{"id":"2105.11686","title":"Towards Understanding the Condensation of Neural Networks at Initial Training","created":"2021-05-25","cited_by_count":32,"area":null},{"id":"2211.06506","title":"Spectral Evolution and Invariance in Linear-width Neural Networks","created":"2022-11-11","cited_by_count":30,"area":null},{"id":"2208.04924","title":"On the Activation Function Dependence of the Spectral Bias of Neural Networks","created":"2022-08-09","cited_by_count":28,"area":null},{"id":"2201.11729","title":"Implicit Regularization in Hierarchical Tensor Factorization and Deep Convolutional Neural Networks","created":"2022-01-27","cited_by_count":27,"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":"2109.09304","title":"Deformed semicircle law and concentration of nonlinear random matrices for ultra-wide neural networks","created":"2021-09-20","cited_by_count":21,"area":null},{"id":"2105.00303","title":"RATT: Leveraging Unlabeled Data to Guarantee Generalization","created":"2021-05-01","cited_by_count":20,"area":null},{"id":"2401.07187","title":"A Survey on Statistical Theory of Deep Learning: Approximation, Training Dynamics, and Generative Models","created":"2024-01-14","cited_by_count":20,"area":null},{"id":"2101.01152","title":"Provable Generalization of SGD-trained Neural Networks of Any Width in the Presence of Adversarial Label Noise","created":"2021-01-04","cited_by_count":17,"area":null},{"id":"2112.05611","title":"Eigenspace Restructuring: a Principle of Space and Frequency in Neural Networks","created":"2021-12-10","cited_by_count":17,"area":null},{"id":"2310.06982","title":"Data Distillation Can Be Like Vodka: Distilling More Times For Better Quality","created":"2023-10-10","cited_by_count":17,"area":null},{"id":"2308.09543","title":"Latent State Models of Training Dynamics","created":"2023-08-18","cited_by_count":16,"area":null},{"id":"2206.10816","title":"Fighting Fire with Fire: Avoiding DNN Shortcuts through Priming","created":"2022-06-22","cited_by_count":15,"area":null},{"id":"2512.20607","title":"Saddle-to-Saddle Dynamics Explains A Simplicity Bias Across Neural Network Architectures","created":"2025-12-23","cited_by_count":15,"area":null},{"id":"2210.04643","title":"Critical Learning Periods for Multisensory Integration in Deep Networks","created":"2022-10-06","cited_by_count":14,"area":null},{"id":"2206.03688","title":"Identifying good directions to escape the NTK regime and efficiently learn low-degree plus sparse polynomials","created":"2022-06-08","cited_by_count":13,"area":null},{"id":"2302.09344","title":"Beyond Distribution Shift: Spurious Features Through the Lens of Training Dynamics","created":"2023-02-18","cited_by_count":11,"area":null},{"id":"2105.05553","title":"Principal Components Bias in Over-parameterized Linear Models, and its Manifestation in Deep Neural Networks","created":"2021-05-12","cited_by_count":10,"area":null},{"id":"2210.04389","title":"DeepMed: Semiparametric Causal Mediation Analysis with Debiased Deep Learning","created":"2022-10-10","cited_by_count":10,"area":null},{"id":"2603.00742","title":"To Use or not to Use Muon: How Simplicity Bias in Optimizers Matters","created":"2026-02-28","cited_by_count":10,"area":null},{"id":"2211.16220","title":"Which Shortcut Solution Do Question Answering Models Prefer to Learn?","created":"2022-11-29","cited_by_count":9,"area":null},{"id":"2201.08082","title":"Kernel Methods and Multi-layer Perceptrons Learn Linear Models in High Dimensions","created":"2022-01-20","cited_by_count":8,"area":null},{"id":"2210.12001","title":"When Expressivity Meets Trainability: Fewer than $n$ Neurons Can Work","created":"2022-10-21","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":"2211.06077","title":"Overparameterized random feature regression with nearly orthogonal data","created":"2022-11-11","cited_by_count":7,"area":null},{"id":"2406.09935","title":"Predicting the Susceptibility of Examples to Catastrophic Forgetting","created":"2024-06-14","cited_by_count":7,"area":null},{"id":"2207.04612","title":"Synergy and Symmetry in Deep Learning: Interactions between the Data, Model, and Inference Algorithm","created":"2022-07-11","cited_by_count":6,"area":null},{"id":"2511.11500","title":"Honesty over Accuracy: Trustworthy Language Models through Reinforced Hesitation","created":"2025-11-14","cited_by_count":5,"area":null},{"id":"2105.00507","title":"Universal scaling laws in the gradient descent training of neural networks","created":"2021-05-02","cited_by_count":4,"area":null},{"id":"2402.05916","title":"GenEFT: Understanding Statics and Dynamics of Model Generalization via Effective Theory","created":"2024-02-08","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":"2106.04010","title":"FEAR: A Simple Lightweight Method to Rank Architectures","created":"2021-06-07","cited_by_count":3,"area":null},{"id":"2202.06054","title":"Towards Data-Algorithm Dependent Generalization: a Case Study on Overparameterized Linear Regression","created":"2022-02-12","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":"2406.17467","title":"Early learning of the optimal constant solution in neural networks and humans","created":"2024-06-25","cited_by_count":3,"area":null},{"id":"2409.02792","title":"UnLearning from Experience to Avoid Spurious Correlations","created":"2024-09-04","cited_by_count":2,"area":"safety"},{"id":"2502.12125","title":"Hypernym Bias: Unraveling Deep Classifier Training Dynamics through the Lens of Class Hierarchy","created":"2025-02-17","cited_by_count":2,"area":null},{"id":"2305.04135","title":"Maintaining Stability and Plasticity for Predictive Churn Reduction","created":"2023-05-06","cited_by_count":1,"area":null},{"id":"2305.15276","title":"Sparse Mean Estimation in Adversarial Settings via Incremental Learning","created":"2023-05-24","cited_by_count":0,"area":null},{"id":"2504.01737","title":"Enlightenment Period Improving DNN Performance","created":"2025-04-02","cited_by_count":0,"area":null},{"id":"2507.20548","title":"Annotation-Free Human Sketch Quality Assessment","created":"2025-07-28","cited_by_count":0,"area":null},{"id":"2602.00576","title":"Data Distribution as a Lever for Guiding Optimizers Toward Superior Generalization in LLMs","created":"2026-01-31","cited_by_count":0,"area":null},{"id":"2603.21169","title":"Model Evolution Under Zeroth-Order Optimization: A Neural Tangent Kernel Perspective","created":"2026-03-22","cited_by_count":0,"area":null},{"id":"2605.04231","title":"Anatomy of a failure: When, how, and why deep vision fails in scientific domains","created":"2026-05-05","cited_by_count":0,"area":null}],"next_cursor":null}