{"total":124,"rows":[{"id":"2209.04836","title":"Git Re-Basin: Merging Models modulo Permutation Symmetries","created":"2022-09-11","cited_by_count":471,"area":null},{"id":"2110.06296","title":"The Role of Permutation Invariance in Linear Mode Connectivity of Neural Networks","created":"2021-10-12","cited_by_count":291,"area":null},{"id":"2408.07666","title":"Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities","created":"2024-08-14","cited_by_count":254,"area":null},{"id":"2305.03053","title":"ZipIt! Merging Models from Different Tasks without Training","created":"2023-05-04","cited_by_count":179,"area":null},{"id":"2410.19735","title":"Model merging with SVD to tie the Knots","created":"2024-10-25","cited_by_count":122,"area":null},{"id":"2211.08403","title":"REPAIR: REnormalizing Permuted Activations for Interpolation Repair","created":"2022-11-15","cited_by_count":120,"area":null},{"id":"2301.12780","title":"Equivariant Architectures for Learning in Deep Weight Spaces","created":"2023-01-30","cited_by_count":107,"area":null},{"id":"2201.09405","title":"Improving Chest X-Ray Report Generation by Leveraging Warm Starting","created":"2022-01-24","cited_by_count":95,"area":null},{"id":"2309.15698","title":"Deep Model Fusion: A Survey","created":"2023-09-27","cited_by_count":91,"area":null},{"id":"2106.15933","title":"Saddle-to-Saddle Dynamics in Deep Linear Networks: Small Initialization Training, Symmetry, and Sparsity","created":"2021-06-30","cited_by_count":85,"area":null},{"id":"2211.08422","title":"Mechanistic Mode Connectivity","created":"2022-11-15","cited_by_count":50,"area":null},{"id":"2306.04251","title":"Stochastic Collapse: How Gradient Noise Attracts SGD Dynamics Towards Simpler Subnetworks","created":"2023-06-07","cited_by_count":39,"area":null},{"id":"2105.14573","title":"Embedding Principle of Loss Landscape of Deep Neural Networks","created":"2021-05-30","cited_by_count":38,"area":null},{"id":"2506.13018","title":"Symmetry in Neural Network Parameter Spaces","created":"2025-06-16","cited_by_count":36,"area":null},{"id":"2210.17216","title":"Symmetries, flat minima, and the conserved quantities of gradient flow","created":"2022-10-31","cited_by_count":35,"area":null},{"id":"2403.01753","title":"Training-Free Pretrained Model Merging","created":"2024-03-04","cited_by_count":32,"area":null},{"id":"2408.11804","title":"Approaching Deep Learning through the Spectral Dynamics of Weights","created":"2024-08-21","cited_by_count":31,"area":null},{"id":"2205.10637","title":"Symmetry Teleportation for Accelerated Optimization","created":"2022-05-21","cited_by_count":25,"area":null},{"id":"2309.16932","title":"Symmetry Induces Structure and Constraint of Learning","created":"2023-09-29","cited_by_count":25,"area":null},{"id":"2403.11871","title":"The Real Tropical Geometry of Neural Networks","created":"2024-03-18","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":"2305.13404","title":"Improving Convergence and Generalization Using Parameter Symmetries","created":"2023-05-22","cited_by_count":23,"area":null},{"id":"2111.15527","title":"Embedding Principle: a hierarchical structure of loss landscape of deep neural networks","created":"2021-11-30","cited_by_count":22,"area":null},{"id":"2307.06966","title":"Layer-wise Linear Mode Connectivity","created":"2023-07-13","cited_by_count":22,"area":null},{"id":"2305.19510","title":"Mildly Overparameterized ReLU Networks Have a Favorable Loss Landscape","created":"2023-05-31","cited_by_count":17,"area":null},{"id":"2402.16828","title":"Training Neural Networks from Scratch with Parallel Low-Rank Adapters","created":"2024-02-26","cited_by_count":17,"area":null},{"id":"2504.09484","title":"An overview of condensation phenomenon in deep learning","created":"2025-04-13","cited_by_count":17,"area":null},{"id":"2305.12133","title":"Loss Spike in Training Neural Networks","created":"2023-05-20","cited_by_count":16,"area":null},{"id":"2403.00986","title":"Merging Text Transformer Models from Different Initializations","created":"2024-03-01","cited_by_count":16,"area":null},{"id":"2502.00264","title":"Beyond the Permutation Symmetry of Transformers: The Role of Rotation for Model Fusion","created":"2025-02-01","cited_by_count":16,"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":"2502.02496","title":"Deep Weight Factorization: Sparse Learning Through the Lens of Artificial Symmetries","created":"2025-02-04","cited_by_count":14,"area":null},{"id":"2411.18607","title":"Task Arithmetic Through The Lens Of One-Shot Federated Learning","created":"2024-11-27","cited_by_count":12,"area":null},{"id":"2502.05300","title":"Parameter Symmetry Potentially Unifies Deep Learning Theory","created":"2025-02-07","cited_by_count":12,"area":null},{"id":"2406.16689","title":"Coding schemes in neural networks learning classification tasks","created":"2024-06-24","cited_by_count":11,"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":"2412.11521","title":"On the Ability of Deep Networks to Learn Symmetries from Data: A Neural Kernel Theory","created":"2024-12-16","cited_by_count":10,"area":null},{"id":"2505.23681","title":"Understanding Mode Connectivity via Parameter Space Symmetry","created":"2025-05-29","cited_by_count":10,"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":"2411.08798","title":"Learning Gaussian Multi-Index Models with Gradient Flow: Time Complexity and Directional Convergence","created":"2024-11-13","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":"2209.10080","title":"Deep Double Descent via Smooth Interpolation","created":"2022-09-21","cited_by_count":8,"area":null},{"id":"2303.13093","title":"Type-II Saddles and Probabilistic Stability of Stochastic Gradient Descent","created":"2023-03-23","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":"2311.01644","title":"Should Under-parameterized Student Networks Copy or Average Teacher Weights?","created":"2023-11-03","cited_by_count":8,"area":null},{"id":"2312.09832","title":"Disentangling Linear Mode-Connectivity","created":"2023-12-15","cited_by_count":8,"area":null},{"id":"2408.15495","title":"Remove Symmetries to Control Model Expressivity and Improve Optimization","created":"2024-08-28","cited_by_count":8,"area":null},{"id":"2410.01452","title":"Ensembles provably learn equivariance through data augmentation","created":"2024-10-02","cited_by_count":8,"area":null},{"id":"2506.02269","title":"A Tale of Two Symmetries: Exploring the Loss Landscape of Equivariant Models","created":"2025-06-02","cited_by_count":8,"area":null},{"id":"2205.13283","title":"Embedding Principle in Depth for the Loss Landscape Analysis of Deep Neural Networks","created":"2022-05-26","cited_by_count":7,"area":null},{"id":"2211.11623","title":"Linear Stability Hypothesis and Rank Stratification for Nonlinear Models","created":"2022-11-21","cited_by_count":7,"area":null},{"id":"2402.01342","title":"Improving Model Fusion by Training-time Neuron Alignment with Fixed Neuron Anchors","created":"2024-02-02","cited_by_count":7,"area":null},{"id":"2402.02362","title":"Unification of Symmetries Inside Neural Networks: Transformer, Feedforward and Neural ODE","created":"2024-02-04","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":"2305.05089","title":"Functional Equivalence and Path Connectivity of Reducible Hyperbolic Tangent Networks","created":"2023-05-08","cited_by_count":6,"area":null},{"id":"2402.05626","title":"Loss Landscape of Shallow ReLU-like Neural Networks: Stationary Points, Saddle Escape, and Network Embedding","created":"2024-02-08","cited_by_count":6,"area":null},{"id":"2410.14837","title":"Topological obstruction to the training of shallow ReLU neural networks","created":"2024-10-18","cited_by_count":6,"area":null},{"id":"2605.03601","title":"Most ReLU Networks Admit Identifiable Parameters","created":"2026-05-05","cited_by_count":6,"area":null},{"id":"2307.08921","title":"Optimistic Estimate Uncovers the Potential of Nonlinear Models","created":"2023-07-18","cited_by_count":5,"area":null},{"id":"2305.11417","title":"Exploring the Complexity of Deep Neural Networks through Functional Equivalence","created":"2023-05-19","cited_by_count":4,"area":null},{"id":"2309.00508","title":"Geometry and Local Recovery of Global Minima of Two-layer Neural Networks at Overparameterization","created":"2023-09-01","cited_by_count":4,"area":null},{"id":"2310.10171","title":"On permutation symmetries in Bayesian neural network posteriors: a variational perspective","created":"2023-10-16","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":"2409.05800","title":"Input Space Mode Connectivity in Deep Neural Networks","created":"2024-09-09","cited_by_count":4,"area":null},{"id":"2505.21722","title":"Saddle-To-Saddle Dynamics in Deep ReLU Networks: Low-Rank Bias in the First Saddle Escape","created":"2025-05-27","cited_by_count":4,"area":null},{"id":"2510.09816","title":"A mathematical theory for understanding when abstract representations emerge in neural networks","created":"2025-10-10","cited_by_count":4,"area":null},{"id":"2603.09581","title":"Towards Understanding Adam Convergence on Highly Degenerate Polynomials","created":"2026-03-10","cited_by_count":4,"area":null},{"id":"2405.13721","title":"Connectivity Shapes Implicit Regularization in Matrix Factorization Models for Matrix Completion","created":"2024-05-22","cited_by_count":3,"area":null},{"id":"2409.01410","title":"Dataset Distillation from First Principles: Integrating Core Information Extraction and Purposeful Learning","created":"2024-09-02","cited_by_count":3,"area":null},{"id":"2411.10649","title":"Deep Loss Convexification for Learning Iterative Models","created":"2024-11-16","cited_by_count":3,"area":null},{"id":"2506.13714","title":"Understanding Learning Invariance in Deep Linear Networks","created":"2025-06-16","cited_by_count":3,"area":null},{"id":"2506.14951","title":"Flat Channels to Infinity in Neural Loss Landscapes","created":"2025-06-17","cited_by_count":3,"area":null},{"id":"2603.22030","title":"On the Interplay of Priors and Overparametrization in Bayesian Neural Network Posteriors","created":"2026-03-23","cited_by_count":3,"area":null},{"id":"2305.14122","title":"Transferring Learning Trajectories of Neural Networks","created":"2023-05-23","cited_by_count":2,"area":null},{"id":"2405.17501","title":"Geometry of Critical Sets and Existence of Saddle Branches for Two-layer Neural Networks","created":"2024-05-26","cited_by_count":2,"area":null},{"id":"2502.11362","title":"Teleportation With Null Space Gradient Projection for Optimization Acceleration","created":"2025-02-17","cited_by_count":2,"area":null},{"id":"2503.06181","title":"Make Haste Slowly: A Theory of Emergent Structured Mixed Selectivity in Feature Learning ReLU Networks","created":"2025-03-08","cited_by_count":2,"area":null},{"id":"2505.05143","title":"Sparse Training from Random Initialization: Aligning Lottery Ticket Masks using Weight Symmetry","created":"2025-05-08","cited_by_count":2,"area":null},{"id":"2506.07467","title":"Circumventing Backdoor Space via Weight Symmetry","created":"2025-06-09","cited_by_count":2,"area":null},{"id":"2510.08023","title":"Do We Really Need Permutations? Impact of Model Width on Linear Mode Connectivity","created":"2025-10-09","cited_by_count":2,"area":null},{"id":"2605.25234","title":"On the Epistemic Uncertainty of Overparametrized Neural Networks","created":"2026-05-24","cited_by_count":2,"area":null},{"id":"2505.22578","title":"Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization","created":"2025-05-28","cited_by_count":1,"area":null},{"id":"2508.19151","title":"Saddle Hierarchy in Dense Associative Memory","created":"2025-08-26","cited_by_count":1,"area":null},{"id":"2602.05849","title":"Visualizing the loss landscapes of physics-informed neural networks","created":"2026-02-05","cited_by_count":1,"area":null},{"id":"2603.13467","title":"Learning Not To Interfere: Adapting Experts For Model Merging","created":"2026-03-13","cited_by_count":1,"area":null},{"id":"2604.03068","title":"Escape dynamics and implicit bias of one-pass SGD in overparameterized quadratic networks","created":"2026-04-03","cited_by_count":1,"area":null},{"id":"2605.21324","title":"Stimulus symmetries can confound representational similarity analyses","created":"2026-05-20","cited_by_count":1,"area":null},{"id":"2607.12332","title":"Gradient Flow Dynamics and Implicit Bias of Diagonal Linear Networks under Infinitesimal Initialization","created":"2026-07-14","cited_by_count":1,"area":null},{"id":"2607.18422","title":"PAC--Bayes Bounds on Quotient Parameter Spaces: Geometry-induced Implicit-Bias Priors","created":"2026-07-20","cited_by_count":1,"area":null},{"id":"2202.09276","title":"Geometric Regularization from Overparameterization","created":"2022-02-18","cited_by_count":0,"area":null},{"id":"2204.04348","title":"Neuronal diversity can improve machine learning for physics and beyond","created":"2022-04-09","cited_by_count":0,"area":null},{"id":"2305.13536","title":"Subspace-Configurable Networks","created":"2023-05-22","cited_by_count":0,"area":null},{"id":"2306.02834","title":"Proximity to Losslessly Compressible Parameters","created":"2023-06-05","cited_by_count":0,"area":null},{"id":"2405.15895","title":"Manifold Metric: A Loss Landscape Approach for Predicting Model Performance","created":"2024-05-24","cited_by_count":0,"area":null},{"id":"2406.18035","title":"Local Linear Recovery Guarantee of Deep Neural Networks at Overparameterization","created":"2024-06-26","cited_by_count":0,"area":null},{"id":"2410.03972","title":"Measuring and Controlling Solution Degeneracy across Task-Trained Recurrent Neural Networks","created":"2024-10-04","cited_by_count":0,"area":null},{"id":"2410.12927","title":"Deep Model Merging: The Sister of Neural Network Interpretability -- A Survey","created":"2024-10-16","cited_by_count":0,"area":null},{"id":"2505.12419","title":"Embedding principle of homogeneous neural network for classification problem","created":"2025-05-18","cited_by_count":0,"area":null},{"id":"2505.13582","title":"Uncovering Critical Sets of Deep Neural Networks via Sample-Independent Critical Lifting","created":"2025-05-19","cited_by_count":0,"area":null},{"id":"2506.19609","title":"Beyond Static Models: Hypernetworks for Adaptive and Generalizable Forecasting in Complex Parametric Dynamical Systems","created":"2025-06-24","cited_by_count":0,"area":null}],"next_cursor":"100"}