Every idea extracted from recent arXiv mathematics papers — verified and unverified. Click an idea to open its full card; badges show the empirical verdict.
Replace dense attention over structured object tokens with attention over role-filler tensor-product representations. A learned query specifies both a role and a filler, retrieves objects matching that binding, extracts a target role, and rebinds the extracted filler into an output object.
Replace the usual sum of pairwise modality similarities with a higher-order score based on the coordinatewise Hadamard product of all normalized modality embeddings. For modalities indexed by i=1,...,m, score a tuple using s(x_1,...,x_m)=\omega^\top(\bar g_1(x_1)\odot\cdots\odot\bar g_m(x_m)), where \omega is learned and \odot is coordinatewise multiplication. This adds explicit m-way interactions without concatenating raw features or introducing a joint encoder.
Replace direct logit gradient updates for a simplex-valued neural module with a cascade consisting of a passive LTI filter followed by softmax. The filter can provide useful memory or momentum, but its transfer function is constrained to remain strictly passive, preventing the destabilization mechanism identified for nonpassive higher-order replicator dynamics.
Replace ordinary modality-specific residual fusion with a switched observer whose latent correction depends on the currently available channel. The individual channels are allowed to be insufficient to reconstruct the latent state; stability is enforced over the full switching cycle, so complementary intermittent observations can jointly maintain a stable representation.
Learn a low-dimensional Koopman operator from successive states of an iterative neural system, such as debate agents, recurrent refinement blocks, or diffusion denoising trajectories. Use the magnitude of the subdominant eigenvalue to predict the remaining number of rounds required for disagreement to fall below a target tolerance, and stop computation when the predicted deadline is reached rather than using a fixed round budget.
Replace a square dense projection in a Transformer or MLP with a trainable recursive butterfly matrix. The layer preserves multiscale channel interactions while constraining every complementary row-column block to rank at most k, reducing parameters and enabling recursive structured matrix-vector products. Unlike an arbitrary sparse layer, the construction has an explicit recursive factorization and a quasi-optimal approximation guarantee among matrices with the same butterfly rank.
Replace consecutive or randomly assigned transformed KV coefficients with groups whose variance-volume is approximately equal. Train one equal-size vector-quantizer codebook per group, so a fixed-width cache does not waste its low-rate budget by forcing high-variance and low-variance coordinates into badly mismatched groups. This is a drop-in quantization-layout change that can be applied to keys, values, or both.
Replace an unconstrained deep RoPE attention residual update by a spherical or norm-preserving update whose attention kernel has a known positive floor. Estimate the reversible transverse spectrum of the current attention matrix and choose the residual step size below its explicit Euler stability limit; use the angular token diameter as a runtime contraction monitor.
Use an online estimate of the positive feedback gain among logits, routing probabilities, and representations to adjust the softmax temperature. Increase temperature when the estimated cyclic gain approaches the instability regime, preventing exponential amplification and router collapse without globally weakening all layers.
Implement causal linear attention in chunks and combine chunk summaries with an associative scan instead of carrying the recurrent state through all chunks sequentially. This preserves the exact causal computation while reducing inter-chunk dependency depth from the number of chunks to its logarithm, enabling substantially more GPU parallelism for long-context training and prefill.
Replace the usual random or elementwise-positive linear-attention feature map with a rank-one positive-semidefinite feature map derived from query and key vectors. For normalized inputs, the resulting kernel is the squared inner product, which is nonnegative and gives a geometrically structured interference pattern that is better suited to associative recall than an arbitrary low-rank feature map.
Replace a dense directed attention matrix by a collection of K learned source-to-hub-to-target interactions. Each hub corresponds to a directed biclique, allowing many source tokens to communicate with many target tokens using O(NK) rather than O(N^2) pair interactions. The construction preserves asymmetric information flow and can be initialized from a graph cover of high-attention edges.
Compress each layer's KV tensor with a partial Tucker approximation over token and feature axes, then encode the truncation residual with a rotated uniform quantizer. Select token rank, feature rank, and residual bit-width jointly under a global byte budget, allowing values with flat spectra to receive residual bits while keys may receive more low-rank capacity.
Give every possible pairwise relation its own recurrent latent state, stored in a dictionary keyed by stable node identities, instead of discarding the state whenever the edge is absent from the current graph. At each timestep, active edges retrieve their old state, update it with the current pair features and neighborhood messages, and write it back after message passing. This should improve predictions in dynamic graphs with intermittent interactions and reduce the burden on node states to…
Use a spherical-design codebook and the paper's polar slack factorization to create a nonnegative geometric interaction bias for attention or expert routing. The resulting kernel is generated by a rank-one term and a rank-at-most-d term, and entries close to zero can define a structured sparse mask instead of relying only on learned top-k selection.
Replace the usual JL dimension choice with a task-specific dimension budget based on the paper's asymptotic ranking law. For retrieval, nearest-neighbor search, or compressed attention keys, choose m from a target Kendall correlation and automatically expand the projection when measured ranking quality falls below the target.
Replace dense or single-dilation sparse attention with two sequential sparse attention stages whose offsets form co-prime arithmetic progressions. The first stage mixes tokens separated by multiples of M2, the second by multiples of M1; their composition reaches virtual offsets mM2+nM1, providing many structured long-range interactions from only M1+M2-1 physical offset families. Use causal masking and residual connections so the module can replace a standard transformer attention block without…
Replace ordinary row-softmax attention with a doubly stochastic Sinkhorn attention plan W, and periodically recover a gauge-fixed pairwise cost from W using the exact double-centering identity. Use this recovered cost to initialize or regularize a structured attention score, making the attention geometry identifiable despite arbitrary query and key row and column offsets.
Support conjunction queries over multiple roles without explicitly storing a huge tensor of repeated objects. Represent the required higher-order memory through query-dependent contractions, enabling compositional retrieval with memory that scales linearly in the number of objects.
Replace a quadratic pairwise attention or graph aggregation kernel with a compact, translation-invariant indefinite kernel approximated by signed random Fourier features. The feature map preserves the kernel's negative spectral mass through a diagonal sign matrix, so the resulting linear-time aggregation can represent similarities that ordinary positive-definite random features cannot.
Replace dense graph self-attention with two parallel branches: exact softmax attention only over graph neighbors and a global linear-attention branch that summarizes all nodes through feature-space statistics. A learned node-wise gate interpolates between the branches, allowing locally structured nodes to use sparse attention while retaining a global-information path.
Construct a cheap graph or token-mixing operator by partitioning nodes into k blocks using the bottom nonconstant eigenvectors of P squared, then replacing dense pairwise mixing with conditional averaging inside each block followed by one baseline propagation step. Unlike ordinary spectral clustering, the bottom modes target partitions where block labels are rapidly destroyed by P, producing an aggressively mixing representation layer rather than a community-preserving pooling layer. The…
Replace an explicit attention step by an implicit Euler step that solves a proximal subproblem involving the Hopfield energy. The new state is evaluated inside the softmax self-consistently, which makes the method less sensitive to large step sizes and can prevent explicit attention from overshooting or tunneling between attraction basins.
Replace an arbitrary graph-attention mask with a fractional edge mask lying in the intersection of the spanning-tree polytope and twice the matching polytope. The mask represents a distribution over connected spanning trees while imposing expected degree at most two at every vertex, after which sampled trees can be used for sparse message passing.