betaFeedback

WorksComputer Science

Condensing Law of Security Information

Abstract

Scaling laws indicate that the generalization error of a large language model (LLM) decreases monotonically with model size once data is sufficient, and the corollary that larger models outperform smaller ones is now widely accepted. We argue that this corollary fails in security: decisions there hinge on anomalous inputs, which are long-tailed (e.g., zero-day attacks) and therefore poorly served by likelihood fitting. Worse still, the capability brought about by scale (e.g., inferring the intended whole from a corrupted part) leads a LLM silently repair the anomaly before any downstream security head ever sees it. We formalize this as a tradeoff between two quantities: the cognitive capability K(W)=I(W;P)K(W)=I(W;P) that a model holds about the world distribution, and the security-information retention R(W)=I(Z;S)I(A;S)R(W)=\frac{I(Z;S)}{I(A;S)} that survives in its intermediate representation. We derive three information-theoretic bounds to connect them: an upper bound on restoration accuracy in terms of K(W)K(W), an upper bound on R(W)R(W) in terms of the restoration error, and a Fano lower bound on security-task error in terms of R(W)R(W). These theorems approximately imply that usable security information is a product of how much security information survives and how well it can be extracted, indicating that optimal usable security information can be achieved at a certain model size. We state this as the condensing law of security information: usable security information is unimodal in model size, so security generalization error is U-shaped and is minimized at a finite N⋆N^{\star} rather than decreasing without bound. The experiments on various real-world LLMs validate the condensing law, indicating that there is an optimal model size to solve security issues.

Paper details

vlnc:W_6avyvesw
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
arXiv nonexclusive license
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Computer Science

Not recorded for this paper

View statement
Looking for related papers…