{"total":56,"rows":[{"id":"2503.10622","title":"Transformers without Normalization","created":"2025-03-13","cited_by_count":70,"area":null},{"id":"2509.14128","title":"Canary-1B-v2 & Parakeet-TDT-0.6B-v3: Efficient and High-Performance Models for Multilingual ASR and AST","created":"2025-09-17","cited_by_count":60,"area":null},{"id":"2502.02732","title":"Peri-LN: Revisiting Normalization Layer in the Transformer Architecture","created":"2025-02-04","cited_by_count":34,"area":null},{"id":"2512.01868","title":"The Mean-Field Dynamics of Transformers","created":"2025-12-01","cited_by_count":30,"area":null},{"id":"2510.19093","title":"Weight Decay may matter more than muP for Learning Rate Transfer in Practice","created":"2025-10-21","cited_by_count":20,"area":null},{"id":"2606.16899","title":"Fantastic Pretraining Optimizers and Where to Find Them II: Hyperball Optimization","created":"2026-06-15","cited_by_count":16,"area":null},{"id":"2511.18890","title":"Nemotron-Flash: Towards Latency-Optimal Hybrid Small Language Models","created":"2025-11-24","cited_by_count":15,"area":null},{"id":"2510.12402","title":"Cautious Weight Decay","created":"2025-10-14","cited_by_count":13,"area":null},{"id":"2602.03685","title":"Universal One-third Time Scaling in Learning Peaked Distributions","created":"2026-02-03","cited_by_count":10,"area":null},{"id":"2506.08001","title":"Reparameterized LLM Training via Orthogonal Equivalence Transformation","created":"2025-06-09","cited_by_count":9,"area":null},{"id":"2510.22026","title":"Normalization in Attention Dynamics","created":"2025-10-24","cited_by_count":9,"area":null},{"id":"2503.17760","title":"CODA: Repurposing Continuous VAEs for Discrete Tokenization","created":"2025-03-22","cited_by_count":8,"area":null},{"id":"2510.03871","title":"Optimal Scaling Needs Optimal Norm","created":"2025-10-04","cited_by_count":8,"area":null},{"id":"2505.22014","title":"Learning in Compact Spaces with Approximately Normalized Transformer","created":"2025-05-28","cited_by_count":6,"area":null},{"id":"2512.14697","title":"Spherical Leech Quantization for Visual Tokenization and Generation","created":"2025-12-16","cited_by_count":6,"area":null},{"id":"2410.16926","title":"Pyramid Vector Quantization for LLMs","created":"2024-10-22","cited_by_count":4,"area":null},{"id":"2410.19352","title":"Interpreting Neural Networks through Mahalanobis Distance","created":"2024-10-25","cited_by_count":3,"area":null},{"id":"2603.06922","title":"NerVE: Nonlinear Eigenspectrum Dynamics in LLM Feed-Forward Networks","created":"2026-03-06","cited_by_count":3,"area":null},{"id":"2502.11646","title":"Hyper-SET: Designing Transformers via Hyperspherical Energy Minimization","created":"2025-02-17","cited_by_count":2,"area":null},{"id":"2509.15874","title":"ENSAM: an efficient foundation model for interactive segmentation of 3D medical images","created":"2025-09-19","cited_by_count":2,"area":null},{"id":"2510.16743","title":"Zero-Shot Performance Prediction for Probabilistic Scaling Laws","created":"2025-10-19","cited_by_count":2,"area":null},{"id":"2602.01212","title":"SimpleGPT: Improving GPT via A Simple Normalization Strategy","created":"2026-02-01","cited_by_count":2,"area":null},{"id":"2605.06067","title":"Normalized Architectures are Natively 4-Bit","created":"2026-05-07","cited_by_count":2,"area":null},{"id":"2606.29158","title":"On the Nonlinearity of Learning Rate Scaling for LLM Training","created":"2026-06-28","cited_by_count":2,"area":null},{"id":"2609.19107","title":"How Model Growth, Recursion, and Boundary Operators Influence Scaling Exponents","created":"2026-09-16","cited_by_count":2,"area":null},{"id":"2410.24050","title":"A Mechanistic Study of Transformers Training Dynamics","created":"2024-10-31","cited_by_count":1,"area":null},{"id":"2503.17715","title":"Normalized Matching Transformer","created":"2025-03-22","cited_by_count":1,"area":null},{"id":"2507.17501","title":"DNT: a Deeply Normalized Transformer that can be trained by Momentum SGD","created":"2025-07-23","cited_by_count":1,"area":null},{"id":"2510.05049","title":"KEEP: Integrating Medical Ontologies with Clinical Data for Robust Code Embeddings","created":"2025-10-06","cited_by_count":1,"area":null},{"id":"2511.01600","title":"Lite ENSAM: a lightweight cancer segmentation model for 3D Computed Tomography","created":"2025-11-03","cited_by_count":1,"area":null},{"id":"2601.22095","title":"GeoNorm: Unify Pre-Norm and Post-Norm with Geodesic Optimization","created":"2026-01-29","cited_by_count":1,"area":null},{"id":"2602.14490","title":"Parameter-Efficient Fine-Tuning of LLMs with Mixture of Space Experts","created":"2026-02-16","cited_by_count":1,"area":null},{"id":"2604.01683","title":"Invertible Query-Key Coupling Composes with Attention Mechanisms","created":"2026-04-02","cited_by_count":1,"area":null},{"id":"2605.00265","title":"Polaris: Coupled Orbital Polar Embeddings for Hierarchical Concept Learning","created":"2026-04-30","cited_by_count":1,"area":null},{"id":"2605.04418","title":"Demystifying Manifold Constraints in LLM Pre-training","created":"2026-05-06","cited_by_count":1,"area":null},{"id":"2605.06611","title":"The Structural Origin of Attention Sink: Variance Discrepancy, Super Neurons, and Dimension Disparity","created":"2026-05-07","cited_by_count":1,"area":null},{"id":"2606.11585","title":"Kuramoto Attention: Synchronizing Self-Attention on the Torus","created":"2026-06-10","cited_by_count":1,"area":null},{"id":"2606.14259","title":"Beyond a Single Explanation of the Adam--SGD Gap","created":"2026-06-12","cited_by_count":1,"area":null},{"id":"2508.09148","title":"Motif 2.6B Technical Report","created":"2025-08-02","cited_by_count":0,"area":null},{"id":"2511.07308","title":"Can Stationary Distributions of Scale-Invariant Neural Networks Be Described by the Thermodynamics of an Ideal Gas?","created":"2025-11-10","cited_by_count":0,"area":null},{"id":"2512.12469","title":"Sparse Concept Anchoring for Interpretable and Controllable Neural Representations","created":"2025-12-13","cited_by_count":0,"area":null},{"id":"2512.22423","title":"Bright 4B: Scaling Hyperspherical Learning for Segmentation in 3D Brightfield Microscopy","created":"2025-12-27","cited_by_count":0,"area":null},{"id":"2603.03135","title":"Torus embeddings","created":"2026-03-03","cited_by_count":0,"area":null},{"id":"2603.13180","title":"MXNorm: Reusing MXFP block scales for efficient tensor normalisation","created":"2026-03-13","cited_by_count":0,"area":null},{"id":"2604.23434","title":"When Does Removing LayerNorm Help? Activation Bounding as a Regime-Dependent Implicit Regularizer","created":"2026-04-25","cited_by_count":0,"area":null},{"id":"2605.11007","title":"The Transformer as a Polar State Estimator","created":"2026-05-10","cited_by_count":0,"area":null},{"id":"2605.11850","title":"Constrained Stochastic Spectral Preconditioning Converges for Nonconvex Objectives","created":"2026-05-12","cited_by_count":0,"area":null},{"id":"2605.13262","title":"Chem-GMNet: A Sphere-Native Geometric Transformer for Molecular Property Prediction","created":"2026-05-13","cited_by_count":0,"area":null},{"id":"2607.02252","title":"ArcAD: Anomaly-Rectified Calibration for Cold-Start Supervised Anomaly Detection","created":"2026-07-02","cited_by_count":0,"area":null},{"id":"2608.13156","title":"Rethinking Normalization Placement for LLMs: Post-Norm under Curriculum Depth Growing","created":"2026-08-13","cited_by_count":0,"area":null},{"id":"2608.31036","title":"Normalized Low-Rank Adaptation","created":"2026-08-31","cited_by_count":0,"area":null},{"id":"2609.07017","title":"HyperTransfer: Understanding the Equivalence between Base Optimizer and Hyperball","created":"2026-09-07","cited_by_count":0,"area":null},{"id":"2609.07058","title":"CNsEMD: An Expert-Annotated Multi-Field-Strength MRI Dataset and a Hyperspherical Manifold Network for Multimodal Cranial Nerve Parcellation","created":"2026-09-07","cited_by_count":0,"area":null},{"id":"2609.34493","title":"LLN: Learnable Lens Networks for Parameter-Efficient Long-Horizon Dynamical Prediction","created":"2026-09-28","cited_by_count":0,"area":null},{"id":"2610.14344","title":"On the Depth Transfer of Neural Optimizers via L-Smoothness","created":"2026-10-09","cited_by_count":0,"area":null},{"id":"2610.20069","title":"Trainable at Depth, Solvable at Small Size: What the Loss Surface of a Polynomial Activation Dictates for Designing, Initializing and Training Networks","created":"2026-10-14","cited_by_count":0,"area":null}],"next_cursor":null}