Memory ideas

External memories, state compression, retrieval-style read/write policies and calibrated stopping rules.

2 ideas found

Category: Memory ML: Attention Math: Spectral theory Clear all

Unverified 2026

Transform-Domain Quantized Edge Attention

Compress a trained graph-attention model's edge-dependent logits or gates in the line-graph Fourier domain. Smooth edge values become concentrated in low-frequency coefficients, allowing low-frequency coefficients to retain more precision while high-frequency residuals use fewer bits or are discarded.

Useful6/10

Difficulty6/10

Novelty8/10

Memory Memory Inference speedup Scalability

Paper: Lossy compression of weighted graph adjacency matrices by transform coding arXiv:2607.14834

Unverified 2026

Signed-triangle Nyström lookahead

Replace one-step greedy landmark selection in Nyström attention or kernel compression with a restricted pairwise-lookahead rule. The lookahead is motivated by the paper's explicit obstruction: a signed triangle can make individual column gains exhibit increasing rather than diminishing returns, so the best next column need not belong to the best pair.

Useful5/10

Difficulty5/10

Novelty4/10

Memory Memory Accuracy Scalability

Paper: Nyström Error Beyond $M$-Matrices: A Minimal Diagonally Dominant Obstruction arXiv:2607.19282