Every idea extracted from recent arXiv mathematics papers — verified and unverified. Click an idea to open its full card; badges show the empirical verdict.
Make Laplacian latency attention hardware-friendly by quantizing the exponential decay and row normalization to powers of two. Exponentials and divisions can then be implemented with lookup tables and integer shifts, enabling a low-bit attention kernel for neuromorphic and edge accelerators.
Approximate a long hidden-state or key/value trajectory by a small sum of exponential modes, storing only the mode nodes and vector weights instead of every timestep. Recover the modes from a Hankel matrix through an annihilating polynomial, then reconstruct the trajectory or a compact recurrent state during inference.
Add a neural feature, adapter, or expert block only through the component of its outputs that is orthogonal to the span of all previously installed blocks. Quotient coefficient directions that produce nearly identical outputs with an SVD or pseudoinverse, so the new block contributes intrinsic representational dimensions instead of duplicating old features. The expected benefit is a smaller effective architecture and better-conditioned block expansion.
Implement f(A)b inside a neural network with a short Lanczos recurrence instead of an eigendecomposition or dense matrix-function operation. Use an SPD operator A such as a regularized graph Laplacian or feature covariance matrix, and choose the number of iterations by monitoring successive approximations. For Stieltjes functions, Lanczos is guaranteed to be close to the best vector in the same Krylov subspace.
Replace repeated full-dimensional matrix-exponential or ODE solves in a conditioned continuous-time state-space layer with contour quadrature evaluated in a projection basis. The same reduced basis and contour nodes can serve many conditioning vectors, while shifted reduced resolvents provide a stable and differentiable approximation over a prescribed time window.
Replace ordinary reverse-mode differentiation through a long sequence of stiff relaxation updates with a projected adjoint that separates slow conserved features from rapidly relaxing residual features. The neural layer can use large outer time steps even when its internal relaxation time is very small, while reconstructing only the microscopic gradient component required by the preceding layer.
Use a strictly upper-triangular block operator to represent interactions between ordered layers, experts, or token groups, and compute its inverse exactly with a finite Neumann series. Because nilpotency truncates the series after a known number of blocks, the module avoids an iterative solver while retaining controlled long-range interactions.
Construct a symmetric token affinity matrix and approximate only its spectrally outlying token-mixing modes using a small set of sampled landmark columns. The resulting low-rank operator replaces an \(O(n^2)\) dense mixer by two skinny matrix multiplications, while the paper's residual guarantee predicts that large-magnitude global interaction modes are preserved.
Replace unconstrained low-rank compression of a neural state with an augmented basis that always contains vectors representing known conserved quantities or diagnostically important linear statistics. After each learned transition, project the state back onto the affine constraint set with an exact minimum-norm correction, preventing rank truncation and model error from accumulating in those statistics.
Construct one reference Lanczos basis for a symmetric propagation operator H, then derive several seed-specific spectral responses for Q_a(H)x_0 through finite-band polynomial connectors. With degree-r seeds, each transformed basis vector uses at most 2r+1 neighboring reference basis vectors, avoiding a separate Lanczos factorization for every seed.
Replace an SVD-based pseudoinverse of a learned rectangular matrix with a low-degree polynomial initialization followed by a few Newton–Schulz iterations. The polynomial approximates the inverse Gram operator, while a cheap residual test accepts it only when the iteration is contractive and otherwise selects a conservative transpose-scaled initialization.
Instrument selected neural-network operators with cheap stochastic perturbations and estimate how much their outputs change under finite-precision perturbations. Use the resulting per-operator score to identify unstable kernels and selectively switch them to FP32, compensated accumulation, or a stable reformulation instead of running the entire model at high precision.
Replace a large stable linear state-space or recurrent layer by a lower-order balanced realization computed from frequency-targeted controllability and observability Gramians. Use generalized low-rank ADI with imaginary-axis shifts concentrated at frequencies that dominate the training data, then retain states associated with the largest approximate Hankel singular values. This should reduce recurrent inference cost while preserving the layer's input-output response in the selected frequency…
Turn stochastic edge thinning into a mass-aware graph sparsifier: first form retained connected blobs, then connect blob summaries using the rank-one shortcut probability from the paper. Choose the retention rate below the explicit subcritical threshold so the contracted graph remains a collection of small decorations around leading blobs, reducing message-passing cost without creating an artificial giant component.
Replace a dense tensor-product positional encoding or first MLP layer with a hierarchical sparse-grid B-spline feature map. The network evaluates only localized basis functions indexed by multi-levels with bounded total level, reducing feature count while retaining high-order approximation for functions with mixed derivative regularity. The basis can initially be fixed and later fine-tuned jointly with the downstream network.
Replace a dense block of a large neural-network weight matrix with an adaptively constructed cross approximation, but prevent premature termination using residual checks on O(m+n) additional diagonal-like entries. Accept a rank only after the probe residual has remained below tolerance for several consecutive iterations; otherwise continue adding pivots.
Replace a dense token-mixing matrix in a sequence model with a fixed or learnable SBP derivative operator D=P^{-1}Q. The discrete integration-by-parts identity makes the interior mixing energy-neutral or boundary-dissipative, reducing exploding activations in deep residual stacks while preserving directional information along the sequence.
Replace a large dense layer whose input and output dimensions factor into multiple modes by a TT-matrix whose parameters are stored as a chain of small cores. Periodically apply TT-SVD rounding to remove weak singular directions and keep the representation within a prescribed approximation error. This transfers the paper's central computational principle—perform tensor-product contractions directly in compressed form—to neural network layers.
Replace a dense recurrent transition matrix with a periodic CMV-style product of alternating local 2x2 unitary cores. The transition is exactly norm-preserving, has O(n) trainable parameters under periodic tying, and can be applied through local factor operations rather than stored as an n-by-n matrix. Use turnover refactorization when changing the ordering or boundary connection of cores, enabling a compact cyclic unitary state-space layer.
Replace a dense weight matrix by a cross approximation built from selected rows and columns rather than by a conventional truncated SVD. Periodically refresh the selected indices using residual leverage scores, warm-starting from the previous factorization so that the compressed layer can track weight changes during fine-tuning.
Add a fixed or weakly parameterized residual mixer whose interaction between sequence positions at distance \(r\) is proportional to \(1/(r\log^2 r)\). Instead of truncating the kernel at a short radius, represent its heavy tail with dyadic distance bands and compute each band using prefix sums or block pooling, giving every token access to arbitrarily distant context at roughly \(O(L\log L)\) cost.
Represent every mesh interface degree of freedom by one feature copy per incident cell, and apply local neural blocks directly to these cell tensors. Enforce inter-cell consistency with valence-weighted averaging only after selected layers or hierarchy transitions, avoiding repeated construction of a global sparse graph or assembled feature vector. This is suited to adaptive quadtrees, octrees, and finite-element neural operators.
Add a conservative correction after low-rank tensor compression so selected linear moments of an activation or learned state are exactly preserved. This can reduce tensor rank and memory without allowing compression error to accumulate in physically meaningful global quantities.
Replace dense grid tokens or global spectral features with coefficients of compactly supported kernels centered on a nested hierarchy of spatial points. Encode an input field into coarse-to-fine coefficients, apply a neural map to those coefficients, and decode the predicted coefficients at arbitrary query locations; the contribution from each level provides an explicit multiscale output decomposition.