B. Patro

The proposed Riemannian--Lorentz Parameter Fusion (RLPF) method projects aligned groups to common coordinates, lifts selected coordinates to the Lorentz hyperboloid model of hyperbolic space, computes a regularized geodesic barycenter, and decodes the result into the two branches.

QT is introduced, a Quantum-inspired Transformer for vision tasks with three components: angle-inspired encoding that maps image tokens to learned trigonometric Hilbert-space features analogous to quantum rotation-based state encoding; self-attention over these periodic features, inducing a classical cosine kernel approximated to quantum fidelity kernels; and gated multiplicative emulation, a trainable classical surrogate for interaction terms found in variational circuits.

Vision models can ditch the scanning: HAMSA achieves state-of-the-art accuracy among SSMs while being 2.2x faster than transformers and using significantly less memory and energy.

Achieve state-of-the-art structure-preserving conditional image generation with diffusion models using only 8M parameters, or even *fewer* parameters than existing adapters.