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 ordinary per-channel or per-token KV quantization with a structured orthogonal transform followed by blockwise 3-bit quantization. Use a normalized Walsh-Hadamard transform and small SO(4) rotations to spread outliers across coordinates, quantize the transformed vectors, and exploit orthogonality to rotate queries and attention outputs so unquantized attention remains mathematically equivalent.
Replace an ordinary group-L1 penalty on structured neural components with a powered ratio-of-norms penalty applied to their nonnegative importance magnitudes. The ratio encourages importance to concentrate on a small number of heads, channels, or experts while being less sensitive to arbitrary rescaling of the underlying weights. After training, components with small importance can be physically removed and the model can be fine-tuned.
Represent a nonnegative attention or routing score matrix by two nonnegative vectors, X = uv^T, and learn only entries on a sparse bipartite graph of important query-key or token-expert interactions. Complete the remaining entries multiplicatively and monitor cycle residuals as a certificate of whether the sparse representation is compatible with rank one. Use local ratio violations to trigger additional edges or relax the rank-one approximation only where needed.