Unverified
2026
Replace scalar entropy penalties on attention maps with a matrix-valued heat-flow regularizer over a circular or periodic token coordinate. Each position stores a positive semidefinite matrix describing coupled heads, experts, or channels; heat smoothing is constrained by the sharp modified log-Sobolev and Bogoliubov–Kubo–Mori contraction rather than an arbitrary smoothing coefficient. This should suppress high-frequency routing noise while preserving positive matrix structure and reducing…
Useful6/10
Difficulty6/10
Novelty7/10
Unverified
2026
Replace ordinary Jacobian penalties in coordinate MLPs or deformation networks with a learned local rotation frame and a polyconvex energy of the relative stretch. Penalize \(U\), its cofactor, and its determinant through a convex function, while separately smoothing the rotation field through \(R^T\operatorname{Curl}R\). The intended benefit is resistance to fold formation and better conditioning than directly penalizing \(\|J-I\|^2\), especially for large deformations.
Useful6/10
Difficulty6/10
Novelty6/10
✓✓ Beats tuned baseline
2026
Replace one deterministic residual update with a short cyclic composition of learned vector fields evaluated for randomized, short run times. Because finite compositions of noncommuting flows generate directional-derivative and Lie-bracket terms, changing the cycle order gives the network an explicit, low-cost way to learn drift directions that are unavailable from the individual vector fields alone.
Useful6/10
Difficulty5/10
Novelty7/10
✗ Failed on benchmark
2026
Regularize a circular recurrent kernel by directly controlling the growth rate and phase velocity of its Fourier modes. This converts replay-speed selection into a low-dimensional spectral control problem and can suppress unstable or excessively slow modes without adding recurrent parameters.
Useful6/10
Difficulty6/10
Novelty8/10
Unverified
2026
Replace a static top-k MoE capacity rule with a router whose expert allocation evolves through a finite-domain coverage process. Experts with larger current occupancy can either receive more future capacity, intentionally amplifying specialization, or receive less capacity by reversing the size dependence, allowing a controlled test of the paper's asymmetry-amplification mechanism.
Useful5/10
Difficulty6/10
Novelty6/10
Unverified
2026
Add a differentiable penalty that encourages a neural implicit field to have a controlled local homogeneity degree across concentric spatial scales. The penalty compares the flux-normalized frequency at adjacent radii, optionally targeting a desired degree k, so the network is discouraged from producing scale-inconsistent or oscillatory local geometry.
Useful5/10
Difficulty5/10
Novelty7/10
Unverified
2026
Use the Gaussian approximation of a high-dimensional maximum to set a simultaneous coordinate-clipping threshold for minibatch gradients or activations. The threshold is sampled from a correlated Gaussian with the observed batch covariance, rather than treating coordinates as independent or estimating an unstable extreme quantile directly.
Useful5/10
Difficulty6/10
Novelty7/10
Unverified
2026
Treat the number K of minibatches between expensive control updates as a review period: the controlled neural dynamics use parameters or decisions computed at time nK and hold them fixed until (n+1)K. Scan K, estimate first and second finite differences of validation loss or episodic return, and use the resulting nonmonotone-to-convex or concave phase diagram to select an update frequency rather than assuming that more frequent updates are always better.
Useful5/10
Difficulty4/10
Novelty6/10
Unverified
2026
Replace or augment a low-dimensional recurrent transition with affine maps whose linear parts belong to a structured unipotent holonomy family, and train the cell so that positive accumulated translation produces a controlled projective attractor. This creates a measurable two-basin long-horizon behavior: hidden-state perturbation directions should align with a learned direction X or its antipode according to the sign of a scalar functional, rather than exhibiting unconstrained rotation or…
Useful5/10
Difficulty6/10
Novelty8/10
Unverified
2026
Insert a fixed reversible lattice shear into a residual network so successive blocks follow a structured monodromy orbit rather than using unrelated learned transformations. Apply the transformation to a small learned subspace of hidden channels while leaving the remaining channels unchanged. This creates deterministic phase-dependent feature mixing with no additional trainable parameters.
Useful5/10
Difficulty5/10
Novelty8/10
Unverified
2026
Replace the fixed decay coefficient of a stochastic recurrent or state-space layer by an adaptive mean-reversion coefficient driven by the cumulative squared hidden-state energy. The controller approximates conditioning the latent trajectory on a small L2 norm: high-energy trajectories receive stronger restoring drift, whereas low-energy trajectories retain the base dynamics and noise.
Useful5/10
Difficulty5/10
Novelty6/10
Unverified
2026
Give a shared neural dynamical state multiple local readout operators, such as a site channel and a neighboring-pair channel, and measure their space-time responses separately. Add a loss that encourages each channel to have its own dominant propagation velocity while constraining every channel to remain inside a common maximum-speed cone. This transfers the paper's result that spectroscopic selection rules reveal complementary dynamical pathways that are invisible in a single response function.
Useful5/10
Difficulty6/10
Novelty7/10
Unverified
2026
For a neural module that forms causal or statistical ratios from minibatch covariances, replace raw denominator penalties and raw-scale uncertainty weights with a log-denominator or relative-error objective. The front-door covariance minor has variance proportional to its squared magnitude, so a small denominator is not intrinsically evidence of poor estimation under the Gaussian model. This should prevent the network from spuriously avoiding valid representations merely because their…
Useful5/10
Difficulty4/10
Novelty7/10
Unverified
2026
For a neural ODE or physics-informed neural network whose residual cancellation is reliable only after temporal averaging, add an analytic temporal corrector that integrates the zero-mean part of the residual over each time cell. The corrector vanishes at cell boundaries and is smaller by a factor of the cell duration, so it improves pointwise-in-time residuals without changing the learned state at synchronization times.
Useful5/10
Difficulty4/10
Novelty7/10
Unverified
2026
Replace a deterministic mixture-of-experts residual block with K population-indexed stochastic expert states coupled through a graphon matrix. The layer uses a shared drift and expert-dependent diffusion, while an empirical convex-order penalty makes later representations more dispersed than a reference representation without permitting a mean shift.
Useful5/10
Difficulty5/10
Novelty7/10
Unverified
2026
Add a weak regularizer that keeps categorical representations away from both uniformity and deterministic collapse by targeting an empirically selected information-variance level. Unlike entropy maximization, this objective does not reward the uniform distribution, because information-content variance is exactly zero at uniformity.
Useful5/10
Difficulty3/10
Novelty6/10
Unverified
2026
Regularize hidden activations or per-example gradients with a discrete version of the paper's Z_E^2 norm. Apply an E-norm to the largest fraction of coordinates and an L2 norm to the remaining tail, allowing the model to preserve a few large responses while discouraging widespread heavy-tailed noise.
Useful5/10
Difficulty3/10
Novelty7/10
Unverified
2026
Use a pressure objective to select expert-routing distributions by balancing task reward against route entropy, rather than optimizing task loss alone. The resulting router behaves like an equilibrium-state estimator: it should retain multiple high-performing branches when their combined entropy outweighs the advantage of a single branch.
Useful5/10
Difficulty5/10
Novelty5/10
Unverified
2026
Replace a time-homogeneous recurrent update by a sequence of parameterized maps f_t, and regularize late-time pairs of updates to approximately commute: applying block f_t followed by f_r should agree with applying f_r followed by f_t. This should make long-horizon predictions robust to local time-step reorderings and schedule perturbations, while proximal statistics provide a diagnostic for whether trajectories repeatedly approach one another rather than diverging permanently.
Useful5/10
Difficulty5/10
Novelty7/10
Unverified
2026
Use the lifted convex hull as a training-time regularizer for pairs of nonnegative neural features, encouraging their empirical second- and third-order interaction statistics to lie in the paper's moment cone. This constrains correlations, squares, and cubic cross-moments jointly through PSD inequalities instead of merely penalizing large activations.
Useful5/10
Difficulty4/10
Novelty8/10
Unverified
2026
Replace an ordinary elementwise interaction between two feature matrices by a noncommutative functional-calculus layer \(\varphi(A,B)\), where \(A\) and \(B\) are Hermitian channel operators that need not commute. Add a soft penalty on \([A,B]=AB-BA\), and use a Besov-smooth parameterization of \(\varphi\) so that perturbations are controlled in Schatten \(p\)-norm for \(p\leq2\). This creates a principled matrix interaction module that can remain stable when feature operators or graph…
Useful5/10
Difficulty6/10
Novelty8/10
Unverified
2026
Replace ordinary isotropic residual noise in a normalized continuous-depth block with projected Brownian forcing on the unit sphere. Apply a shared random symmetric quadratic drift to all tokens, plus a small token-specific tangent perturbation; the shared term preserves structured antipodal dynamics while the independent term removes persistent symmetry and cluster degeneracy. This is intended as a controlled stochastic regularizer, not merely additive Gaussian noise.
Useful5/10
Difficulty5/10
Novelty7/10
Unverified
2026
Use spatially correlated training points whose low-frequency structure factor vanishes instead of iid points. For neural fields, PINNs, image-coordinate MLPs, or spatially indexed minibatches, this should suppress long-wavelength quadrature and gradient-estimation noise while preserving the represented target dynamics. The finite-order prediction is that a design with structure factor S(k)=O(|k|^{2q}) produces lower variance for smooth losses than iid sampling, especially as the domain or batch…
Useful5/10
Difficulty5/10
Novelty6/10
Unverified
2026
Replace raw updates of strongly coupled parameter blocks by updates in rescaled, approximately normal-form coordinates. The optimizer estimates the local coupling matrix between block directions, solves a small modulation system for transformed velocities, and optionally subtracts predictable first-order cross-block drift.
Useful5/10
Difficulty5/10
Novelty4/10