pith. sign in

Latent-space Attacks for Refusal Evasion in Language Models

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

1 Pith paper citing it

fields

cs.AI 1

years

2026 1

verdicts

UNVERDICTED 1

clear filters

representative citing papers

Latent-space Attacks for Refusal Evasion in Language Models

cs.AI · 2026-05-20 · unverdicted · novelty 6.0 · 2 refs

Refusal suppression via difference-in-means ablation equals projection onto a linear probe's decision boundary, and a controlled evasion attack optimizing confidence past the boundary achieves SOTA success rates on 15 models.

citing papers explorer

Showing 1 of 1 citing paper after filters.

  • Latent-space Attacks for Refusal Evasion in Language Models cs.AI · 2026-05-20 · unverdicted · none · ref 24 · 2 links

    Refusal suppression via difference-in-means ablation equals projection onto a linear probe's decision boundary, and a controlled evasion attack optimizing confidence past the boundary achieves SOTA success rates on 15 models.