✓✓ Beats tuned baseline
2026
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.
Useful7/10
Difficulty4/10
Novelty5/10
△ Mechanism confirmed, baseline not beaten
2026
Replace Muon's single momentum matrix with a weighted mixture of fast and slow relaxation modes. The fast mode tracks rapidly changing gradients while the slow mode preserves a longer-horizon direction; their mixture is semi-orthogonalized and applied as the matrix update.
Useful7/10
Difficulty4/10
Novelty6/10
△ Mechanism confirmed, baseline not beaten
2026
Replace full PSD self-attention with a pivoted Cholesky/Nyström approximation whose landmarks are sampled from the unexplained diagonal mass. Tokens with large residual self-similarity are more likely to become landmarks, so the rank budget is spent on difficult regions rather than uniformly selected tokens.
Useful7/10
Difficulty5/10
Novelty5/10
△ Mechanism confirmed, baseline not beaten
2026
Train one prompt-conditioned controller to solve a distribution of stochastic control tasks directly from the control objective, instead of generating an optimal trajectory dataset for every task. Use the probability-flow velocity to evolve particles deterministically, evaluate running and terminal costs on those particles, and backpropagate through the rollout to learn a reusable operator.
Useful7/10
Difficulty6/10
Novelty6/10
△ Mechanism confirmed, baseline not beaten
2026
Replace simultaneous parameter updates with sequential block updates whose order is selected using estimated cross-block sensitivity. The paper shows that sequential policy updates can have a substantially smaller local contraction factor than decoupled or differently ordered updates; the neural analogue is to order attention, normalization, backbone, and head blocks according to the spectral radius of their composed update map.
Useful7/10
Difficulty6/10
Novelty7/10
✗ Failed on benchmark
2026
Insert a linear Johnson–Lindenstrauss bottleneck around a set of jointly processed representations, choosing its width from the sharp finite-set dimension bound rather than from the model's nominal hidden size. The projection should preserve pairwise distances between tokens, patches, or retrieved items, allowing a downstream attention or MLP block to operate at lower width while retaining the geometry relevant to similarity computations.
Useful7/10
Difficulty5/10
Novelty5/10
△ Mechanism confirmed, baseline not beaten
2026
Replace multiplicative query-key attention scores with an affinity based on the l1 distance between first-spike latency vectors. For each query token and key token, small latency differences produce large affinity and distant timings decay exponentially, yielding a locality-sensitive attention pattern naturally compatible with leaky spiking neurons.
Useful7/10
Difficulty5/10
Novelty7/10
✗ Mechanism failed
2026
Train a small encoder and latent Koopman predictor to forecast whether a neural sequence model will enter a high-error or high-instability region, then execute an expensive refinement block only when the forecasted risk exceeds a threshold. The base model remains active at every step, so the learned preview model controls computation rather than directly replacing the main predictor. Add a bounded-rate interpolation when the gate switches off, preventing abrupt changes in recurrent state or…
Useful7/10
Difficulty5/10
Novelty7/10
✗ Failed on benchmark
2026
Replace the usual momentum schedule in a neural-network optimizer with a discretization of the paper's lemniscate-acceleration ODE. The method uses a time-dependent friction coefficient that is initially very large and then decays according to lemniscate sine and cosine functions, targeting faster reduction of the gradient norm than constant-momentum SGD or standard Nesterov schedules.
Useful7/10
Difficulty5/10
Novelty8/10
△ Mechanism confirmed, baseline not beaten
2026
Construct a bipartite candidate graph between tokens and experts from the router’s top-k logits, then solve a capacity-constrained maximum-cardinality matching rather than dispatching each token independently. The mechanism targets the extreme tail of routing completion: it should reduce unmatched or repeatedly reassigned tokens and lower maximum dispatch delay and expert starvation, even when average routing quality changes little.
Useful7/10
Difficulty6/10
Novelty4/10
△ Mechanism confirmed, baseline not beaten
2026
Compress an existing dense neural-network weight matrix into a recursive butterfly operator using Gaussian sketches of complementary blocks. This is useful for deployment or distillation: the dense model provides an oracle for matrix-vector products, while the compressed model stores only recursive transfer bases and small cores. The generalized Nyström identity gives exact reconstruction for rank-k blocks and a principled approximation route for numerically low-rank blocks.
Useful7/10
Difficulty7/10
Novelty6/10
✓✓ Beats tuned baseline
2026
Replace dense token-to-token attention by a multiscale spiderweb communication pattern. Tokens first aggregate upward through a dyadic hierarchy, communicate horizontally only with a small number of cells at the appropriate height, and then receive information broadcast downward. Hyperbolic distance supplies a principled rule for choosing the height at which two tokens interact: nearby tokens interact at fine scales, while far-apart tokens interact through coarse representatives.
Useful7/10
Difficulty5/10
Novelty6/10
△ Mechanism confirmed, baseline not beaten
2026
Replace the exact matrix-polar normalization in Muon with the smoothed feedback \(h_\epsilon(M)=M(M^\top M+\epsilon I)^{-1/2}\). This retains singular-vector-aware updates and approximately unit-normalizes dominant spectral modes, but avoids unstable behavior when the momentum matrix is rank deficient or has tiny singular values.
Useful7/10
Difficulty5/10
Novelty4/10
△ Mechanism confirmed, baseline not beaten
2026
Use the relaxed QFT tensor-network topology as a trainable norm-preserving mixer inside a neural block, replacing a dense token-mixing matrix or an expensive global convolution. The network learns data-adapted global interactions while retaining structured O(N log^2 N) application and an exact cheap inverse, making it suitable for image tokens, long sequences, or reversible residual blocks.
Useful7/10
Difficulty6/10
Novelty6/10
✗ Failed on benchmark
2026
Replace a static batch size with a schedule optimized for a prescribed learning-rate schedule and a fixed total number of processed examples. Steps whose stochastic-gradient noise has a large effect on the paper's loss bound receive larger batches, with the weighting determined by the future learning-rate tail rather than by a hand-designed warmup or cooldown rule.
Useful7/10
Difficulty5/10
Novelty5/10
△ Mechanism confirmed, baseline not beaten
2026
Replace dense attention between tokens on opposite sides of a one-dimensional boundary or segment split with a dyadic low-rank approximation of a Cauchy/Hankel distance kernel. Each distance-scale block uses O(log(1/\varepsilon)) features, and the number of active scales grows only logarithmically with context length after discarding a narrow boundary layer. This is especially suitable for a relative-position attention branch or state-space-like long-range branch, rather than arbitrary…
Useful7/10
Difficulty5/10
Novelty6/10
△ Mechanism confirmed, baseline not beaten
2026
Replace a portion of quadratic key-value attention or an external episodic table with a per-sample matrix fast memory updated by rank-one delta corrections. The memory directly learns a linear key-to-value map and can be carried across sequence segments, providing cheap online adaptation with constant state size per head.
Useful7/10
Difficulty5/10
Novelty5/10
✗ Failed on benchmark
2026
Replace a neural-network optimizer's globally fixed learning-rate geometry with an adaptive quadratic trust region. At every update, construct a local curvature model, accept or reject the step using the ratio between realized and predicted loss decrease, and expand or contract the radius accordingly; the same controller should automatically become conservative in nonconvex regions and Newton-like near a well-conditioned minimum.
Useful7/10
Difficulty6/10
Novelty6/10
✗ Mechanism failed
2026
Replace Lorentz-hyperboloid tensors with proper-velocity tensors whose spatial coordinates can be transformed by standard Euclidean affine layers and activations. Reconstruct the Lorentz time coordinate only at manifold boundaries, preserving the hyperbolic representation while avoiding repeated projection, normalization, or fragile exponential-map calculations.
Useful7/10
Difficulty4/10
Novelty6/10
✗ Failed on benchmark
2026
Construct overlapping attention windows from a block design instead of using one dense sequence-by-sequence attention matrix. Every token appears in exactly $r$ windows and every token pair co-occurs in exactly $\lambda$ windows, giving uniform coverage and avoiding the uneven connectivity of arbitrary sparse masks.
Useful7/10
Difficulty5/10
Novelty6/10
✗ Failed on benchmark
2026
Replace a capacity-penalty-only MoE router with a nonnegative shadow price for each expert, capacity bucket, or hardware resource. Route each token using predicted utility minus the relevant price, while computing a decomposed optimistic objective that certifies how much utility remains above the feasible routed value.
Useful7/10
Difficulty4/10
Novelty5/10
△ Mechanism confirmed, baseline not beaten
2026
Grow a neural network by appending a trainable block together with an analytically initialized inverse block, so the newly added depth is exactly the identity at insertion time. After insertion, untie and optimize the two blocks independently; this preserves the current function while providing additional trainable degrees of freedom. For architectures with one expensive mixing operation followed by cheap channelwise blocks, the same construction can increase depth without repeatedly paying for…
Useful7/10
Difficulty5/10
Novelty5/10
✗ Failed on benchmark
2026
Compress a transformer KV cache by selecting actual past tokens whose key or hidden-state columns form a stable basis for all cached tokens. Instead of retaining tokens with the largest attention scores or leverage scores independently, compute rank-revealing pivoting of the leading right-singular-vector matrix and retain its pivot columns, then evaluate attention using the representatives plus an optional low-cost residual correction.
Useful7/10
Difficulty5/10
Novelty6/10
△ Mechanism confirmed, baseline not beaten
2026
Replace dense query-key attention with an adaptive cross approximation constructed from selected query and key pivot tokens. At each rank, choose the pivot pair that removes large estimated residual energy, update the residual by a rank-1 cross correction, and stop when the residual estimate reaches a target tolerance. The resulting factorization computes approximate attention using a small number of landmark interactions while adapting to the actual token distribution.
Useful7/10
Difficulty6/10
Novelty5/10