Develops a multiplier-based framework via linear programming to prove energy dissipation for IMEX-LMMs up to order 8 on gradient flows, with first results claimed for BDF6 and orders above 6.
Title resolution pending
14 Pith papers cite this work. Polarity classification is still indexing.
14
Pith papers citing it
citation-role summary
background 2
method 1
citation-polarity summary
years
2026 14verdicts
UNVERDICTED 14representative citing papers
Introduces T-Hamiltonian and T-symplectic tensors and derives a constructive T-Williamson normal form for tensors whose Fourier-domain slices are real symmetric positive-definite matrices.
Derives ODE limits of Adam-DA showing that first- and second-order momentum parameters reverse their convergence roles in zero-sum games compared to minimization, validated on GAN experiments.
Neural tangent kernel from neural reparameterization modulates sensitivity and wave tangent kernels to produce spectral filtering, wavenumber modulation, and frequency bias that improve NeurFWI convergence.
Identifiability is proven for recurrent nonlinear switching dynamical systems under flexible assumptions, and ΩSDS is introduced as a flow-based estimator that improves disentanglement and forecasting over VAE-based methods.
ResRL decouples shared semantics between positive and negative responses in LLM reinforcement learning via SVD-based projection residuals, outperforming baselines including NSR by up to 9.4% on math reasoning benchmarks.
TF-LLMER resolves optimization barriers in LLM-enhanced recommenders through embedding normalization and Rec-PCA that aligns semantic representations with collaborative co-occurrence graphs.
MOMENT is a moment-based method for selecting and estimating parameters in multiresponse linear mixed-effects models, with claimed finite-sample consistency under sub-Weibull errors.
Introduces HRC model for game-theoretic decomposition of preferences into orthogonal transitive and cyclic components, paired with DSPPO for dynamic Nash-seeking alignment, reporting gains over BT and GPM baselines on RewardBench and downstream LLM evaluations.
A Neyman-orthogonal moment estimator with adjusted nonparametric fixed effects achieves root-NT asymptotic normality for common parameters in two-way heterogeneous panel models.
SPHERE applies a Parseval penalty to MoE policies in continual RL to maintain spectral plasticity, yielding 133% and 50% higher average success on MetaWorld and HumanoidBench versus unregularized MoE baselines.
Extends alternating numerical scheme from full-dimensional to elliptic lower-dimensional quasi-periodic solutions with a priori bounds, demonstrated on Hénon-Heiles and FPU models using multi-scale analysis and resolvent gluing.
The paper defines L2 lateral string stability using an arc-length perspective and shows that onboard sensing cannot guarantee error attenuation while V2V communication can.
An inner-outer iteration algorithm with optimal parameters is developed for the stochastic Lyapunov matrix equation, with convergence proven under mean-square stability assumptions and demonstrated via numerical examples.
citing papers explorer
-
Energy Dissipation Analysis of Implicit-Explicit Linear Multistep Methods for Gradient Flows Using General Multipliers
Develops a multiplier-based framework via linear programming to prove energy dissipation for IMEX-LMMs up to order 8 on gradient flows, with first results claimed for BDF6 and orders above 6.
-
Hamiltonian and Symplectic Tensors in the T-product Algebra
Introduces T-Hamiltonian and T-symplectic tensors and derives a constructive T-Williamson normal form for tensors whose Fourier-domain slices are real symmetric positive-definite matrices.
-
Understanding Dynamics of Adam in Zero-Sum Games: An ODE Approach
Derives ODE limits of Adam-DA showing that first- and second-order momentum parameters reverse their convergence roles in zero-sum games compared to minimization, validated on GAN experiments.
-
Deciphering Neural Reparameterized Full-Waveform Inversion with Neural Sensitivity Kernel and Wave Tangent Kernel
Neural tangent kernel from neural reparameterization modulates sensitivity and wave tangent kernels to produce spectral filtering, wavenumber modulation, and frequency bias that improve NeurFWI convergence.
-
End-to-End Identifiable and Consistent Recurrent Switching Dynamical Systems
Identifiability is proven for recurrent nonlinear switching dynamical systems under flexible assumptions, and ΩSDS is introduced as a flow-based estimator that improves disentanglement and forecasting over VAE-based methods.
-
ResRL: Boosting LLM Reasoning via Negative Sample Projection Residual Reinforcement Learning
ResRL decouples shared semantics between positive and negative responses in LLM reinforcement learning via SVD-based projection residuals, outperforming baselines including NSR by up to 9.4% on math reasoning benchmarks.
-
Break the Optimization Barrier of LLM-Enhanced Recommenders: A Theoretical Analysis and Practical Framework
TF-LLMER resolves optimization barriers in LLM-enhanced recommenders through embedding normalization and Rec-PCA that aligns semantic representations with collaborative co-occurrence graphs.
-
Moment-Based Selection of Multiresponse Linear Mixed-Effects Models
MOMENT is a moment-based method for selecting and estimating parameters in multiresponse linear mixed-effects models, with claimed finite-sample consistency under sub-Weibull errors.
-
Transitivity Meets Cyclicity: Explicit Preference Decomposition for Dynamic Large Language Model Alignment
Introduces HRC model for game-theoretic decomposition of preferences into orthogonal transitive and cyclic components, paired with DSPPO for dynamic Nash-seeking alignment, reporting gains over BT and GPM baselines on RewardBench and downstream LLM evaluations.
-
Inference on Linear Regressions with Two-Way Unobserved Heterogeneity
A Neyman-orthogonal moment estimator with adjusted nonparametric fixed effects achieves root-NT asymptotic normality for common parameters in two-way heterogeneous panel models.
-
SPHERE: Mitigating the Loss of Spectral Plasticity in Mixture-of-Experts for Deep Reinforcement Learning
SPHERE applies a Parseval penalty to MoE policies in continual RL to maintain spectral plasticity, yielding 133% and 50% higher average success on MetaWorld and HumanoidBench versus unregularized MoE baselines.
-
Numerical Construction of Elliptic Lower-Dimensional Quasi-Periodic Solutions with a Priori Bound
Extends alternating numerical scheme from full-dimensional to elliptic lower-dimensional quasi-periodic solutions with a priori bounds, demonstrated on Hénon-Heiles and FPU models using multi-scale analysis and resolvent gluing.
-
Lateral String Stability for Vehicle Platoons: Formulation, Definition, and Analysis
The paper defines L2 lateral string stability using an arc-length perspective and shows that onboard sensing cannot guarantee error attenuation while V2V communication can.
-
An Inner-Outer Iteration Algorithm with Optimal Parameters for Stochastic Lyapunov Matrix Equation
An inner-outer iteration algorithm with optimal parameters is developed for the stochastic Lyapunov matrix equation, with convergence proven under mean-square stability assumptions and demonstrated via numerical examples.