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.
Title resolution pending
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
years
2026 8verdicts
UNVERDICTED 8roles
background 2polarities
background 2representative citing papers
Proposes pointwise Riemannian Dimension from feature eigenvalues to derive tighter, representation-aware generalization bounds for deep networks in the nonlinear regime.
CAWI replaces standard random initialization of input-to-hidden weights in randomized neural networks with samples drawn from a data-fitted copula that preserves observed feature dependencies, yielding consistent accuracy gains on 83 classification benchmarks.
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.
MAS-PNCG accelerates IPC by incrementally updating multilevel MAS preconditioners via Sparse-Input Woodbury, adding Hessian-aware 2D subspace minimization and per-subdomain CCD, achieving up to 5.66x speedup over Newton-PCG baselines.
In LVLMs, attention can be replaced by random Gaussian weights with little or no performance loss, indicating that current models get lost in attention rather than efficiently using visual context.
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.
A GPSOL- and DERL-based adaptive controller for first-order SISO nonlinear systems derives LMI-based peak-to-peak gains to set prediction error rates that enforce user-defined output error bounds, verified in simulation and on a pneumatic rig.
citing papers explorer
-
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.
-
Pointwise Generalization in Deep Neural Networks
Proposes pointwise Riemannian Dimension from feature eigenvalues to derive tighter, representation-aware generalization bounds for deep networks in the nonlinear regime.
-
CAWI: Copula-Aligned Weight Initialization for Randomized Neural Networks
CAWI replaces standard random initialization of input-to-hidden weights in randomized neural networks with samples drawn from a data-fitted copula that preserves observed feature dependencies, yielding consistent accuracy gains on 83 classification benchmarks.
-
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.
-
An Efficient Multilevel Preconditioned Nonlinear Conjugate Gradient Method for Incremental Potential Contact
MAS-PNCG accelerates IPC by incrementally updating multilevel MAS preconditioners via Sparse-Input Woodbury, adding Hessian-aware 2D subspace minimization and per-subdomain CCD, achieving up to 5.66x speedup over Newton-PCG baselines.
-
Large Vision-Language Models Get Lost in Attention
In LVLMs, attention can be replaced by random Gaussian weights with little or no performance loss, indicating that current models get lost in attention rather than efficiently using visual context.
-
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.
-
Online Learning-Based Control with Guaranteed Error Bounds for a Class of Nonlinear Systems
A GPSOL- and DERL-based adaptive controller for first-order SISO nonlinear systems derives LMI-based peak-to-peak gains to set prediction error rates that enforce user-defined output error bounds, verified in simulation and on a pneumatic rig.