Pith. sign in

Title resolution pending

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

stat.ML 1

years

2025 1

verdicts

CONDITIONAL 1

representative citing papers

How Private is Your Attention? Bridging Privacy with In-Context Learning

stat.ML · 2025-04-22 · conditional · novelty 6.0

Differentially private pretraining of linear attention heads for in-context linear regression has excess risk that decays like 1/(N L^3) in low dimensions and D^2/(N^2 L^2) in high dimensions, up to log factors and privacy parameters.

citing papers explorer

Showing 1 of 1 citing paper.

  • How Private is Your Attention? Bridging Privacy with In-Context Learning stat.ML · 2025-04-22 · conditional · none · ref 2

    Differentially private pretraining of linear attention heads for in-context linear regression has excess risk that decays like 1/(N L^3) in low dimensions and D^2/(N^2 L^2) in high dimensions, up to log factors and privacy parameters.