Imagen de Portada

A Minimal Bifurcation Model of Load Imbalance in a Softmax Mixture-of-Experts Router

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autor principal:	Kiselev, O. M.
Formato:	Preprint
Publicado:	2026
Materias:	Dynamical Systems Artificial Intelligence Machine Learning
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Manifold Trajectories in Next-Token Prediction: From Replicator Dynamics to Softmax Equilibrium
por: Lee-Jenkins, Christopher R.
Publicado: (2025)

Least-Loaded Expert Parallelism: Load Balancing An Imbalanced Mixture-of-Experts
por: Nguyen, Xuan-Phi, et al.
Publicado: (2026)

Confabulation dynamics in a reservoir computer: Filling in the gaps with untrained attractors
por: O'Hagan, Jack, et al.
Publicado: (2025)

Dynamical Systems Theory Behind a Hierarchical Reasoning Model
por: Es'kin, Vasiliy A., et al.
Publicado: (2026)

Large-Step Training Dynamics of a Two-Factor Linear Transformer Model
por: Balasubramanian, Krishnakumar
Publicado: (2026)

Position: Why a Dynamical Systems Perspective is Needed to Advance Time Series Modeling
por: Durstewitz, Daniel, et al.
Publicado: (2026)

A Hybrid Approach of Transfer Learning and Physics-Informed Modeling: Improving Dissolved Oxygen Concentration Prediction in an Industrial Wastewater Treatment Plant
por: Koksal, Ece S., et al.
Publicado: (2024)

The Recurrent Sticky Hierarchical Dirichlet Process Hidden Markov Model
por: Słupiński, Mikołaj, et al.
Publicado: (2024)

On the Limitations of Fractal Dimension as a Measure of Generalization
por: Tan, Charlie B., et al.
Publicado: (2024)

Memory in Plain Sight: Surveying the Uncanny Resemblances of Associative Memories and Diffusion Models
por: Hoover, Benjamin, et al.
Publicado: (2023)

A Mixture of Experts Gating Network for Enhanced Surrogate Modeling in External Aerodynamics
por: Nabian, Mohammad Amin, et al.
Publicado: (2025)

A Mathematical Conflict Framework for Contextual Data Modulation
por: Kartal, Hakan Emre
Publicado: (2026)

Identifying Equivalent Training Dynamics
por: Redman, William T., et al.
Publicado: (2023)

A-PINN: Auxiliary Physics-informed Neural Networks for Structural Vibration Analysis in Continuous Euler-Bernoulli Beam
por: Saini, Shivani, et al.
Publicado: (2025)

Hybrid Energy-Based Models for Physical AI: Provably Stable Identification of Port-Hamiltonian Dynamics
por: Betteti, Simone, et al.
Publicado: (2026)

Horizon-Constrained Rashomon Sets for Chaotic Forecasting
por: Kale, Gauri, et al.
Publicado: (2026)

True Zero-Shot Inference of Dynamical Systems Preserving Long-Term Statistics
por: Hemmer, Christoph Jürgen, et al.
Publicado: (2025)

Out-of-Domain Generalization in Dynamical Systems Reconstruction
por: Göring, Niclas, et al.
Publicado: (2024)

Optimal Recurrent Network Topologies for Dynamical Systems Reconstruction
por: Hemmer, Christoph Jürgen, et al.
Publicado: (2024)

Inverse Depth Scaling From Most Layers Being Similar
por: Liu, Yizhou, et al.
Publicado: (2026)

Late Fusion Neural Operators for Extrapolation Across Parameter Space in Partial Differential Equations
por: van Tegelen, Eva, et al.
Publicado: (2026)

Language models as master equation solvers
por: Liu, Chuanbo, et al.
Publicado: (2023)

Detecting Invariant Manifolds in ReLU-Based RNNs
por: Eisenmann, Lukas, et al.
Publicado: (2025)

Decomposing heterogeneous dynamical systems with graph neural networks
por: Allier, Cédric, et al.
Publicado: (2024)

Improved deep learning of chaotic dynamical systems with multistep penalty losses
por: Chakraborty, Dibyajyoti, et al.
Publicado: (2024)

Bayesian Inference in Recurrent Explicit Duration Switching Linear Dynamical Systems
por: Słupiński, Mikołaj, et al.
Publicado: (2024)

On the dimension of pullback attractors in recurrent neural networks
por: Fadera, Muhammed
Publicado: (2025)

Inverse Approximation Theory for Nonlinear Recurrent Neural Networks
por: Wang, Shida, et al.
Publicado: (2023)

Sparse identification of nonlinear dynamics and Koopman operators with Shallow Recurrent Decoder Networks
por: Gao, Mars Liyao, et al.
Publicado: (2025)

How to Tame Your LLM: Semantic Collapse in Continuous Systems
por: Wyss, C. M.
Publicado: (2025)

Routers in Vision Mixture of Experts: An Empirical Study
por: Liu, Tianlin, et al.
Publicado: (2024)

A Hybrid Adaptive Velocity Aided Navigation Filter with Application to INS/DVL Fusion
por: Or, Barak, et al.
Publicado: (2022)

LongSSM: On the Length Extension of State-space Models in Language Modelling
por: Wang, Shida
Publicado: (2024)

GRIP: Algorithm-Agnostic Machine Unlearning for Mixture-of-Experts via Geometric Router Constraints
por: Zhu, Andy, et al.
Publicado: (2026)

Graph Mixture of Experts and Memory-augmented Routers for Multivariate Time Series Anomaly Detection
por: Huang, Xiaoyu, et al.
Publicado: (2024)

When is a System Discoverable from Data? Discovery Requires Chaos
por: Shumaylov, Zakhar, et al.
Publicado: (2025)

Kernel Dynamics under Path Entropy Maximization
por: Das, Jnaneshwar
Publicado: (2026)

Learning Chaotic Systems and Long-Term Predictions with Neural Jump ODEs
por: Krach, Florian, et al.
Publicado: (2024)

StableSSM: Alleviating the Curse of Memory in State-space Models through Stable Reparameterization
por: Wang, Shida, et al.
Publicado: (2023)

SymFlux: deep symbolic regression of Hamiltonian vector fields
por: Evangelista-Alvarado, M. A., et al.
Publicado: (2025)