{"total":188,"rows":[{"id":"2109.01903","title":"Robust fine-tuning of zero-shot models","created":"2021-09-04","cited_by_count":797,"area":null},{"id":"2209.04836","title":"Git Re-Basin: Merging Models modulo Permutation Symmetries","created":"2022-09-11","cited_by_count":471,"area":null},{"id":"2103.10427","title":"The Low-Rank Simplicity Bias in Deep Networks","created":"2021-03-18","cited_by_count":154,"area":null},{"id":"2102.02887","title":"Do We Actually Need Dense Over-Parameterization? In-Time Over-Parameterization in Sparse Training","created":"2021-02-04","cited_by_count":110,"area":null},{"id":"2204.03044","title":"Fusing finetuned models for better pretraining","created":"2022-04-06","cited_by_count":110,"area":null},{"id":"2309.15698","title":"Deep Model Fusion: A Survey","created":"2023-09-27","cited_by_count":91,"area":null},{"id":"2202.00661","title":"When Do Flat Minima Optimizers Work?","created":"2022-02-01","cited_by_count":78,"area":null},{"id":"2211.13609","title":"PAC-Bayes Compression Bounds So Tight That They Can Explain Generalization","created":"2022-11-24","cited_by_count":76,"area":null},{"id":"2102.13042","title":"Loss Surface Simplexes for Mode Connecting Volumes and Fast Ensembling","created":"2021-02-25","cited_by_count":75,"area":null},{"id":"2205.12411","title":"Linear Connectivity Reveals Generalization Strategies","created":"2022-05-24","cited_by_count":56,"area":null},{"id":"2402.02364","title":"Loss Landscape Degeneracy and Stagewise Development in Transformers","created":"2024-02-04","cited_by_count":40,"area":null},{"id":"2201.08025","title":"Low-Pass Filtering SGD for Recovering Flat Optima in the Deep Learning Optimization Landscape","created":"2022-01-20","cited_by_count":39,"area":null},{"id":"2209.14764","title":"Model Zoos: A Dataset of Diverse Populations of Neural Network Models","created":"2022-09-29","cited_by_count":39,"area":null},{"id":"2203.09678","title":"Self-Ensemble Adversarial Training for Improved Robustness","created":"2022-03-18","cited_by_count":38,"area":null},{"id":"2312.04339","title":"Merging by Matching Models in Task Parameter Subspaces","created":"2023-12-07","cited_by_count":36,"area":null},{"id":"2306.13253","title":"Predicting Grokking Long Before it Happens: A look into the loss landscape of models which grok","created":"2023-06-23","cited_by_count":35,"area":null},{"id":"2107.11228","title":"Taxonomizing local versus global structure in neural network loss landscapes","created":"2021-07-23","cited_by_count":34,"area":null},{"id":"2211.05641","title":"Regression as Classification: Influence of Task Formulation on Neural Network Features","created":"2022-11-10","cited_by_count":29,"area":null},{"id":"2104.11044","title":"Analyzing Monotonic Linear Interpolation in Neural Network Loss Landscapes","created":"2021-04-22","cited_by_count":28,"area":null},{"id":"2106.16004","title":"What can linear interpolation of neural network loss landscapes tell us?","created":"2021-06-30","cited_by_count":28,"area":null},{"id":"2107.05802","title":"How many degrees of freedom do we need to train deep networks: a loss landscape perspective","created":"2021-07-13","cited_by_count":28,"area":null},{"id":"2108.01289","title":"AdvRush: Searching for Adversarially Robust Neural Architectures","created":"2021-08-03","cited_by_count":28,"area":null},{"id":"2301.12618","title":"ForkMerge: Mitigating Negative Transfer in Auxiliary-Task Learning","created":"2023-01-30","cited_by_count":28,"area":null},{"id":"2101.09048","title":"Selfish Sparse RNN Training","created":"2021-01-22","cited_by_count":26,"area":null},{"id":"2108.01070","title":"Finding Discriminative Filters for Specific Degradations in Blind Super-Resolution","created":"2021-08-02","cited_by_count":26,"area":null},{"id":"2206.08684","title":"Sparse Double Descent: Where Network Pruning Aggravates Overfitting","created":"2022-06-17","cited_by_count":26,"area":null},{"id":"2310.19103","title":"Proving Linear Mode Connectivity of Neural Networks via Optimal Transport","created":"2023-10-29","cited_by_count":25,"area":null},{"id":"2203.16462","title":"Convergence of gradient descent for deep neural networks","created":"2022-03-30","cited_by_count":24,"area":null},{"id":"2209.07509","title":"Random initialisations performing above chance and how to find them","created":"2022-09-15","cited_by_count":23,"area":null},{"id":"2203.14661","title":"Random matrix analysis of deep neural network weight matrices","created":"2022-03-28","cited_by_count":21,"area":null},{"id":"2312.11875","title":"Sparse is Enough in Fine-tuning Pre-trained Large Language Models","created":"2023-12-19","cited_by_count":21,"area":null},{"id":"2405.20231","title":"The Empirical Impact of Neural Parameter Symmetries, or Lack Thereof","created":"2024-05-30","cited_by_count":21,"area":null},{"id":"2110.08616","title":"GradSign: Model Performance Inference with Theoretical Insights","created":"2021-10-16","cited_by_count":19,"area":null},{"id":"2302.12250","title":"Phase diagram of early training dynamics in deep neural networks: effect of the learning rate, depth, and width","created":"2023-02-23","cited_by_count":19,"area":null},{"id":"2211.03959","title":"Pretraining in Deep Reinforcement Learning: A Survey","created":"2022-11-08","cited_by_count":18,"area":null},{"id":"2405.16350","title":"A Second-Order Perspective on Model Compositionality and Incremental Learning","created":"2024-05-25","cited_by_count":18,"area":null},{"id":"2507.12224","title":"Optimizers Qualitatively Alter Solutions And We Should Leverage This","created":"2025-07-16","cited_by_count":17,"area":null},{"id":"2208.13219","title":"Visualizing high-dimensional loss landscapes with Hessian directions","created":"2022-08-28","cited_by_count":16,"area":null},{"id":"2104.07639","title":"Robust Optimization for Multilingual Translation with Imbalanced Data","created":"2021-04-15","cited_by_count":15,"area":null},{"id":"2211.16667","title":"Dynamic Sparse Training via Balancing the Exploration-Exploitation Trade-off","created":"2022-11-30","cited_by_count":15,"area":null},{"id":"2510.06213","title":"Training Dynamics Impact Post-Training Quantization Robustness","created":"2025-10-07","cited_by_count":15,"area":null},{"id":"2102.00485","title":"Exploring the Geometry and Topology of Neural Network Loss Landscapes","created":"2021-01-31","cited_by_count":14,"area":null},{"id":"2106.09467","title":"Algorithmic Bias and Data Bias: Understanding the Relation between Distributionally Robust Optimization and Data Curation","created":"2021-06-17","cited_by_count":13,"area":null},{"id":"2406.04284","title":"What is Dataset Distillation Learning?","created":"2024-06-06","cited_by_count":13,"area":null},{"id":"2505.17646","title":"Unveiling the Basin-Like Loss Landscape in Large Language Models","created":"2025-05-23","cited_by_count":13,"area":null},{"id":"2301.00008","title":"Effects of Data Geometry in Early Deep Learning","created":"2022-12-29","cited_by_count":12,"area":null},{"id":"2401.08667","title":"Data-Driven Physics-Informed Neural Networks: A Digital Twin Perspective","created":"2024-01-05","cited_by_count":12,"area":null},{"id":"2309.14597","title":"Policy Optimization in a Noisy Neighborhood: On Return Landscapes in Continuous Control","created":"2023-09-26","cited_by_count":11,"area":null},{"id":"2205.15142","title":"Special Properties of Gradient Descent with Large Learning Rates","created":"2022-05-30","cited_by_count":10,"area":null},{"id":"2210.01019","title":"Plateau in Monotonic Linear Interpolation -- A \"Biased\" View of Loss Landscape for Deep Networks","created":"2022-10-03","cited_by_count":10,"area":null},{"id":"2302.05823","title":"Data efficiency and extrapolation trends in neural network interatomic potentials","created":"2023-02-12","cited_by_count":10,"area":null},{"id":"2310.14227","title":"Revisiting Deep Ensemble for Out-of-Distribution Detection: A Loss Landscape Perspective","created":"2023-10-22","cited_by_count":10,"area":null},{"id":"2404.06498","title":"Simultaneous linear connectivity of neural networks modulo permutation","created":"2024-04-09","cited_by_count":10,"area":null},{"id":"2407.13954","title":"Neural topology optimization: the good, the bad, and the ugly","created":"2024-07-19","cited_by_count":10,"area":null},{"id":"2105.01303","title":"Personalized Algorithm Generation: A Case Study in Learning ODE Integrators","created":"2021-05-04","cited_by_count":9,"area":null},{"id":"2110.15288","title":"Hyper-Representations: Self-Supervised Representation Learning on Neural Network Weights for Model Characteristic Prediction","created":"2021-10-28","cited_by_count":9,"area":null},{"id":"2111.10489","title":"Modeling Design and Control Problems Involving Neural Network Surrogates","created":"2021-11-20","cited_by_count":9,"area":null},{"id":"2206.03927","title":"Boundary between noise and information applied to filtering neural network weight matrices","created":"2022-06-08","cited_by_count":9,"area":null},{"id":"2206.11120","title":"Near-optimal control of dynamical systems with neural ordinary differential equations","created":"2022-06-22","cited_by_count":9,"area":null},{"id":"2402.09240","title":"Switch EMA: A Free Lunch for Better Flatness and Sharpness","created":"2024-02-14","cited_by_count":9,"area":null},{"id":"2404.06391","title":"Exploring Neural Network Landscapes: Star-Shaped and Geodesic Connectivity","created":"2024-04-09","cited_by_count":9,"area":null},{"id":"2407.05385","title":"Harmony in Diversity: Merging Neural Networks with Canonical Correlation Analysis","created":"2024-07-07","cited_by_count":9,"area":null},{"id":"2504.02666","title":"BECAME: BayEsian Continual Learning with Adaptive Model MErging","created":"2025-04-03","cited_by_count":9,"area":null},{"id":"2506.13234","title":"The Butterfly Effect: Neural Network Training Trajectories Are Highly Sensitive to Initial Conditions","created":"2025-06-16","cited_by_count":9,"area":null},{"id":"2206.12041","title":"How many labelers do you have? A closer look at gold-standard labels","created":"2022-06-24","cited_by_count":8,"area":null},{"id":"2303.00046","title":"Edit at your own risk: evaluating the robustness of edited models to distribution shifts","created":"2023-02-28","cited_by_count":8,"area":null},{"id":"2509.11348","title":"On Linear Mode Connectivity of Mixture-of-Experts Architectures","created":"2025-09-14","cited_by_count":8,"area":"moe"},{"id":"2102.13233","title":"Spurious Local Minima Are Common for Deep Neural Networks with Piecewise Linear Activations","created":"2021-02-25","cited_by_count":7,"area":null},{"id":"2110.06018","title":"On the Security Risks of AutoML","created":"2021-10-12","cited_by_count":7,"area":null},{"id":"2211.12092","title":"Linear Interpolation In Parameter Space is Good Enough for Fine-Tuned Language Models","created":"2022-11-22","cited_by_count":7,"area":null},{"id":"2305.07845","title":"Understanding and Improving Model Averaging in Federated Learning on Heterogeneous Data","created":"2023-05-13","cited_by_count":7,"area":null},{"id":"2306.11922","title":"No Wrong Turns: The Simple Geometry Of Neural Networks Optimization Paths","created":"2023-06-20","cited_by_count":7,"area":null},{"id":"2307.08436","title":"DOT: A Distillation-Oriented Trainer","created":"2023-07-17","cited_by_count":7,"area":"distill"},{"id":"2311.17094","title":"In Search of a Data Transformation That Accelerates Neural Field Training","created":"2023-11-28","cited_by_count":7,"area":null},{"id":"2406.16300","title":"Landscaping Linear Mode Connectivity","created":"2024-06-24","cited_by_count":7,"area":null},{"id":"2206.13034","title":"Monitoring Shortcut Learning using Mutual Information","created":"2022-06-27","cited_by_count":6,"area":null},{"id":"2210.09658","title":"ROSE: Robust Selective Fine-tuning for Pre-trained Language Models","created":"2022-10-18","cited_by_count":6,"area":null},{"id":"2605.18632","title":"Position: Weight Space Should Be a First-Class Generative AI Modality","created":"2026-05-18","cited_by_count":6,"area":null},{"id":"2111.15133","title":"LossPlot: A Better Way to Visualize Loss Landscapes","created":"2021-11-30","cited_by_count":5,"area":null},{"id":"2501.15718","title":"CENSOR: Defense Against Gradient Inversion via Orthogonal Subspace Bayesian Sampling","created":"2025-01-27","cited_by_count":5,"area":null},{"id":"2503.12649","title":"FW-Merging: Scaling Model Merging with Frank-Wolfe Optimization","created":"2025-03-16","cited_by_count":5,"area":null},{"id":"2602.04417","title":"EMA Policy Gradient: Taming Reinforcement Learning for LLMs with EMA Anchor and Top-k KL","created":"2026-02-04","cited_by_count":5,"area":"lr"},{"id":"2107.03770","title":"Federated Learning as a Mean-Field Game","created":"2021-07-08","cited_by_count":4,"area":null},{"id":"2203.03692","title":"Low-Loss Subspace Compression for Clean Gains against Multi-Agent Backdoor Attacks","created":"2022-03-07","cited_by_count":4,"area":"safety"},{"id":"2205.06304","title":"Overparameterization Improves StyleGAN Inversion","created":"2022-05-12","cited_by_count":4,"area":null},{"id":"2207.05666","title":"Zero-shot Cross-lingual Transfer is Under-specified Optimization","created":"2022-07-12","cited_by_count":4,"area":null},{"id":"2302.08571","title":"A Comprehensive Review and a Taxonomy of Edge Machine Learning: Requirements, Paradigms, and Techniques","created":"2023-02-16","cited_by_count":4,"area":null},{"id":"2306.17630","title":"Navigating Noise: A Study of How Noise Influences Generalisation and Calibration of Neural Networks","created":"2023-06-30","cited_by_count":4,"area":null},{"id":"2309.09240","title":"High-dimensional manifold of solutions in neural networks: insights from statistical physics","created":"2023-09-17","cited_by_count":4,"area":null},{"id":"2309.16369","title":"Bringing the Discussion of Minima Sharpness to the Audio Domain: a Filter-Normalised Evaluation for Acoustic Scene Classification","created":"2023-09-28","cited_by_count":4,"area":null},{"id":"2402.03187","title":"How Good is a Single Basin?","created":"2024-02-05","cited_by_count":4,"area":null},{"id":"2407.20724","title":"Exploring Loss Landscapes through the Lens of Spin Glass Theory","created":"2024-07-30","cited_by_count":4,"area":null},{"id":"2408.09140","title":"Learning to Explore for Stochastic Gradient MCMC","created":"2024-08-17","cited_by_count":4,"area":null},{"id":"2410.20035","title":"Training the Untrainable: Introducing Inductive Bias via Representational Alignment","created":"2024-10-26","cited_by_count":4,"area":null},{"id":"2105.12920","title":"Search Spaces for Neural Model Training","created":"2021-05-27","cited_by_count":3,"area":null},{"id":"2208.04803","title":"Exploring the trade off between human driving imitation and safety for traffic simulation","created":"2022-08-09","cited_by_count":3,"area":"driving"},{"id":"2210.06819","title":"Mean-field analysis for heavy ball methods: Dropout-stability, connectivity, and global convergence","created":"2022-10-13","cited_by_count":3,"area":null},{"id":"2211.15596","title":"A survey of deep learning optimizers -- first and second order methods","created":"2022-11-28","cited_by_count":3,"area":null},{"id":"2302.04810","title":"Machine Learning Systems: A Survey from a Data-Oriented Perspective","created":"2023-02-09","cited_by_count":3,"area":null},{"id":"2310.10006","title":"Soft ascent-descent as a stable and flexible alternative to flooding","created":"2023-10-16","cited_by_count":3,"area":null}],"next_cursor":"100"}