pith. sign in

Exploring the Limits of Model-Targeted Indiscriminate Data Poisoning Attacks

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

1 Pith paper citing it

fields

cs.LG 1

years

2026 1

verdicts

UNVERDICTED 1

representative citing papers

When Stronger Triggers Backfire: A High-Dimensional Theory of Backdoor Attacks

cs.LG · 2026-05-21 · unverdicted · novelty 8.0

In the proportional high-dimensional regime, stronger backdoor training triggers improve clean accuracy and make attack success non-monotonic for regularized GLMs on Gaussian mixtures, with closed-form proofs for squared loss and fixed-point extensions to convex losses.

citing papers explorer

Showing 1 of 1 citing paper.

  • When Stronger Triggers Backfire: A High-Dimensional Theory of Backdoor Attacks cs.LG · 2026-05-21 · unverdicted · none · ref 28

    In the proportional high-dimensional regime, stronger backdoor training triggers improve clean accuracy and make attack success non-monotonic for regularized GLMs on Gaussian mixtures, with closed-form proofs for squared loss and fixed-point extensions to convex losses.