Prune redundant attention heads using separate similarity scores for sink behavior and content routing. Two heads are considered safely redundant only when their normalized content compositions are close in Aitchison distance and their sink-mass trajectories are also close, avoiding pruning decisions dominated by a shared sink token.
Attach an evidential cost head to a neural graph model, representing each edge cost by a weighted set of interval boxes, and compress this representation before the downstream shortest-path or routing solver. Instead of minimizing Jaccard or Jousselme distance between the original and compressed mass functions, choose merges that minimize the induced cost error on the currently selected route, while enforcing a conservative monotonicity condition so that the resulting path regret is bounded.
Embed every variable-domain sample into one fixed ambient hyperrectangle and append its signed-distance function as an additional channel to the operator input. Deterministically extend fields outside the physical domain, resample them onto a shared latent grid, apply standard Fourier layers, then interpolate and mask the output on the requested target discretization. The network learns the operator rather than a separate geometry encoder, so the same weights can be used across shapes and mesh…
Represent a 3D neural field using high-order polynomial coefficients attached to an adaptively refined tetrahedral mesh, with a small MLP predicting residual corrections from local coordinates. Refine only tetrahedra whose prediction, rendering, or PDE residual is large, and use globally ordered vertices so neighboring tetrahedra share identical face and edge coefficients without hanging-node constraint solves.
Build a recurrent ReLU module that evaluates a binary refinement cascade using a fixed-dimensional state and shared cell weights. Replace hard binary digit selection with two overlapping circle coordinates; switch between their affine state updates only at points where the two candidate updates agree, so the switch is an exact continuous piecewise-linear ReLU operation rather than a multiplicative gate.
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.