Architecture ideas

Research ideas extracted from mathematics papers, categorized as Architecture.

Unverified 2026

WKB-Stokes Sparse Mixer

Replace a dense channel-mixing matrix in a sequence layer with alternating diagonal propagation and sparse unipotent Stokes jumps. The diagonal part carries independently controlled exponential phases, while the unipotent factors implement cheap residual-like mode conversion without changing determinant or requiring a dense matrix multiply. Constrain the phase magnitudes and jump coefficients during training to obtain a reversible, norm-monitorable mixer.

Useful5/10
Difficulty4/10
Novelty7/10
Paper: Holonomy Asymptotics along Quartic Differential Rays arXiv:2608.04729
Unverified 2026

Schur-Complement Barrier Message Passing

Add a small number of latent region-offset variables to a graph or token-mixing layer, interpreting selected edges as low-permeability barriers that suppress cross-region information flow. Eliminate the latent variables analytically, yielding a visible-node update with a structured low-rank correction rather than adding persistent hidden node states. The module is intended to preserve within-cluster propagation while preventing oversmoothing or contamination across learned boundaries.

Useful5/10
Difficulty6/10
Novelty6/10
Paper: An unfitted finite element discrete fracture model for low-permeability barriers via local stiffness matrix modification arXiv:2608.04431
Unverified 2026

Unitary-dilated stochastic router

Replace a softmax transition or mixture-of-experts router by probabilities obtained from squared amplitudes of an isometric latent transition. Each input state is mapped to an orthogonal latent subspace, and summing probability over the latent index produces the desired expert or next-state distribution. The latent amplitudes can retain information that would be destroyed by directly averaging expert outputs, while normalization is guaranteed by construction.

Useful5/10
Difficulty6/10
Novelty5/10
Paper: The Born Representation Theorem and the Unistochastic Theorem arXiv:2608.04354
Unverified 2026

Trace-Free Hodge Feature Mixer

Build a parameter-free spectral channel mixer whose channels are arranged as components of an l-form and whose multiplier is the trace-free Beurling--Ahlfors transform. At every nonzero spatial frequency it mixes the exact and coexact channel subspaces with opposite signs, preventing a uniform channel-direction bias and preserving a structured cancellation property. Insert it as a residual branch before a convolution, MLP, or attention block, with one learned scalar gate controlling its…

Useful5/10
Difficulty6/10
Novelty8/10
Paper: The trace-free Beurling--Ahlfors transform and the Bourgain--Brezis problem for Hodge systems arXiv:2608.04237
Unverified 2026

Entropy-Minimum POVM Router

Replace a scalar softmax classifier or MoE router with a positive-operator-valued measurement computed from learned class or expert density matrices. The resulting operators are positive semidefinite and sum exactly to the identity, so routing probabilities remain normalized for every input state while retaining matrix-valued uncertainty and correlations between latent directions.

Useful5/10
Difficulty6/10
Novelty8/10
Paper: Unifying quantum measurement constructions via a relative-entropy minimum change principle arXiv:2608.04055
Unverified 2026

Noise-aware Walsh-Hadamard bottleneck

Insert a Walsh-Hadamard transform before a quantized categorical or activation bottleneck and assign coordinate-dependent quantization precision using the attenuation spectrum of a quaternary symmetric noise model. Coordinates corresponding to tensor-product frequencies with many nonzero indices are attenuated by higher powers of \(\delta\), so their quantization can be made coarser with little effect on the reconstructed post-noise representation. This creates a structured, fast transform…

Useful5/10
Difficulty4/10
Novelty6/10
Paper: Frequency Coding over Noisy Sampling arXiv:2608.00539
Unverified 2026

Entropy-Certified Interaction Supports

Replace a dense third-order channel-interaction tensor by a fixed sparse support selected through the paper's uniform-marginal infeasibility certificate. Supports with a large dual margin have an effective entropy base below the channel alphabet size, suggesting fewer independent interaction slices and cheaper contractions. Use the certificate either during architecture search or as a pruning score for an already-trained tensorized layer.

Useful5/10
Difficulty5/10
Novelty7/10
Paper: Slice and Partition Rank Criteria for Polynomial Zero-Avoidance arXiv:2607.29490
Unverified 2026

Flow-Constrained Hierarchical Policy

Parameterize a tree-structured policy through realization weights satisfying sequence-form flow conservation, instead of independently predicting probabilities at every node. Conditional action probabilities are recovered by dividing a child sequence weight by its parent weight, guaranteeing globally consistent probabilities and avoiding invalid or contradictory branch masses. This is suitable for hierarchical RL policies, adaptive computation trees, and neural routers with sequential gating…

Useful5/10
Difficulty4/10
Novelty6/10
Paper: Baseball, An Extensive-Form Game-Theoretic Duel arXiv:2607.29041
Unverified 2026

Determinantal Exclusion Router

Replace independent softmax expert choices with a collision-free Markov router whose particles occupy expert positions on a one-dimensional or circular index lattice. A particle can move only to an empty neighboring expert, and the move rate contains a product of sine ratios that globally repels nearby assignments; this should reduce expert collapse and produce more evenly spread routing without requiring a separate pairwise diversity loss.

Useful5/10
Difficulty7/10
Novelty8/10
Paper: Exact Results for the Symmetric Dyson Exclusion Process arXiv:2607.28807
Unverified 2026

Covariance-complement uncertainty head

Represent uncertainty of a graph-structured neural feature field through dual covariance rather than explicitly storing a dense primal covariance. Recover calibrated primal marginal variances from dual statistics using the paper's covariance-complement identity.

Useful5/10
Difficulty4/10
Novelty8/10
Paper: Accelerated Random-Sweep Gibbs Sampling for Gaussian Graphical Models via Dual Normal Factor Graphs arXiv:2607.28706
Unverified 2026

Twisted-Shift Feature Mixer

Build a neural feature-mixing layer from a truncated shift S and a diagonal phase operator T satisfying TS=qST, with |q|=1. The relation forces moving one position in the graded feature basis to multiply the phase operator by q, providing a compact inductive bias for periodic, phase-sensitive, or cyclic data.

Useful5/10
Difficulty4/10
Novelty7/10
Paper: On the diversity of twisted commuting operators arXiv:2607.28372
Unverified 2026

q-Boson Spectral Initialization

Initialize a neural layer with singular values taken from the finite spectral measure of the paper's q-boson Jacobi operator instead of using Xavier or ordinary orthogonal initialization. The resulting layer has a deliberately shaped singular-value distribution and an explicit finite-size spectral edge, allowing initialization to target stable signal propagation while retaining spectral diversity.

Useful5/10
Difficulty4/10
Novelty7/10
Paper: Sharp Bounds on Ground State Energy of the SYK Model arXiv:2607.27185
Unverified 2026

Antichain Multiscale Attention Windows

Replace an unrestricted collection of nested dyadic attention windows on a 2D token grid by a sparse antichain: no selected window may contain another selected window. Use the paper's exponential occupancy guarantee to control how many attention blocks reuse the same token, and add a differentiable log-moment penalty during training when exact antichain selection is relaxed. The expected benefit is bounded peak KV reuse and more predictable sparse-attention cost without discarding multiscale…

Useful5/10
Difficulty5/10
Novelty7/10
Paper: An antichain approach to a conjecture of Zygmund arXiv:2607.25957
Unverified 2026

Central-Moment Feature Mixer

Replace raw polynomial interactions between neighboring feature vectors with central polynomial interactions computed after subtracting the local feature mean. Keep separate second-, third-, and fourth-order channels and apply independent residual gates to them, so a uniform shift of every feature in a neighborhood cannot create artificial cross-order responses. This is a drop-in higher-order mixer for a small transformer or graph neural network.

Useful5/10
Difficulty5/10
Novelty6/10
Paper: Central-Hermite Sensing and Collision for Frame-Robust Order-Resolved Relaxation on D3Q125 arXiv:2607.23629
Unverified 2026

Shared-support heterogeneous-strength adapters

Replace a collection of dense task-specific linear layers with a common sparse structural matrix and task-specific edge strengths. All tasks share the same learned connectivity pattern, but retain independent values on active connections, allowing parameter sharing without forcing identical interactions.

Useful5/10
Difficulty5/10
Novelty5/10
Paper: Estimation of multiple precision matrices under shared support with heterogeneous edge strengths arXiv:2607.23577
Unverified 2026

Metric-magnitude pooling

Replace mean or max pooling over a set of learned element embeddings with pooling based on the metric-magnitude weighting. Pairwise distances create a globally coupled correction for redundancy, so geometrically isolated or boundary elements can contribute differently from dense clusters of nearly duplicate elements.

Useful5/10
Difficulty5/10
Novelty8/10
Paper: Scalably computing metric magnitude arXiv:2607.23354
Unverified 2026

Orthogonal symmetric pair embedding

For every unordered pair of scalar features, construct invariant coordinates from the elementary symmetric quantities s=x+y and q=xy, then feed a truncated orthogonalized polynomial basis in (s,q) to the neural network. Estimate the basis by weighted Gram-Schmidt or Cholesky whitening under the paper's triangle weight, so polynomial channels have low redundancy and controlled scale instead of requiring an unconstrained MLP to learn both symmetry and decorrelation.

Useful5/10
Difficulty3/10
Novelty7/10
Paper: Symmetric Jacobi Polynomials on a Triangle and Their Spectral Algebra arXiv:2607.22751
Unverified 2026

Flux-Balanced Local-Nonlocal Neural Layer

Partition a sequence, image, or graph into regions processed by a cheap local operator and a more expressive nonlocal operator, then couple their boundary activations with a shared continuity equation and a conservative interface-flux equation. The interface correction prevents the local and global branches from creating discontinuities or duplicated information, allowing nonlocal computation to be restricted to selected regions while preserving global consistency.

Useful5/10
Difficulty5/10
Novelty7/10
Paper: Coupling of Local and Nonlocal Problems Using Local Boundary Conditions arXiv:2607.22672
Unverified 2026

Polynomial Spectral Mode Controller

Replace an eigendecomposition-based spectral controller in a small recurrent or state-space transition layer with explicit polynomial projectors. Each hidden state is split into invariant modes, and each mode receives a separately constrained recurrent multiplier, enabling direct suppression of unstable modes or selective retention of long-memory modes using only matrix-polynomial evaluations.

Useful5/10
Difficulty6/10
Novelty6/10
Paper: A Direct Polynomial Approach to Spectral Decomposition arXiv:2607.20218
Unverified 2026

Yang–Baxter Pairwise Router

Replace unconstrained pairwise token-routing interactions with a structured two-token router derived from an involutive set-theoretical Yang–Baxter solution. The pair operator is a convex interpolation between identity and a permutation of discrete routing states, so it cannot amplify probability mass or logits when applied to routing distributions. The Yang–Baxter relation provides a falsifiable test for whether three-token routing updates are insensitive to the two admissible…

Useful5/10
Difficulty6/10
Novelty8/10
Paper: Integrable multi-species SSEP with reactive particle species arXiv:2607.18959
Unverified 2026

Rational-Pole Neural Field Pooling

Replace dense spatial pooling or integral evaluation over a planar domain by a sparse cubature layer whose nodes are poles of a rational approximation fitted only on the domain boundary. For analytic or nearly analytic neural-field channels, the same learned field can then be integrated using substantially fewer evaluations than a uniform grid, while the boundary approximation residual supplies a cheap reliability signal.

Useful5/10
Difficulty6/10
Novelty8/10
Paper: Cubature from rational approximation arXiv:2607.17851
Unverified 2026

Protected-Kernel Graph Diffusion

Replace an ordinary graph diffusion or message-passing operator with a positive-semidefinite Laplacian whose kernel contains a prescribed node-wise subspace. The layer smooths only feature components orthogonal to that subspace, preserving global constants, positional modes, or other structural signals even when graph edges are dynamically added or removed.

Useful5/10
Difficulty5/10
Novelty7/10
Paper: Laplacian Spectral Shaping for Non-Uniform Scaling Formation Control of Open Multi-Agent Systems arXiv:2607.16709
Unverified 2026

Hadamard fractal Fourier encoding

Construct positional features from a self-similar digit system whose Fourier characters are orthogonal under a prescribed nonuniform measure, rather than sampling frequencies independently. Use several admissible multiplier values to create frequency bands while preserving the underlying Hadamard structure, giving a deterministic multiscale encoding with a better-conditioned feature Gram matrix on fractal or highly clustered coordinates.

Useful5/10
Difficulty4/10
Novelty5/10
Paper: Spectral eigenvalue set of self-similar measures associated with product-form Hadamard triples arXiv:2607.15743
Unverified 2026

LU-Preconditioned Orthogonal Weight Retraction

Periodically project a rectangular neural-network weight matrix onto an approximately orthonormal-column matrix using LU-preconditioned CholeskyQR rather than ordinary QR or a polar iteration. Pivoted LU handles badly scaled and nearly dependent columns, while Householder orthogonalization of the LU factor produces a triangular preconditioner that makes the subsequent Cholesky step safer in fp16 or bfloat16.

Useful5/10
Difficulty6/10
Novelty5/10
Paper: RCLUPPr: a new randomized CholeskyQR with LU preconditioning arXiv:2607.15561