betaFeedback

WorksComputer Science

Protecting Safety during Reasoning Training with Low-Rank Fisher Regularisation

Abstract

Large language models are trained to have safety guardrails, but these can be compromised while fine-tuning for new capabilities, such as reasoning. One way to mitigate this degradation is to modify the fine-tuning objective to preserve safety behaviour, but doing so effectively and efficiently remains challenging. We revisit Elastic Weight Consolidation (EWC), which penalises parameter updates based on sensitivity estimates from the diagonal of the Fisher information matrix (FIM). While EWC provides some protection, the diagonal approximation ignores interactions between parameters. We address this limitation by exploiting recent findings that safety-relevant behaviour is located in a low-rank subspace within models. Our method, Subspace-Projected Elastic Weight Consolidation (SP-EWC), projects gradients onto this subspace to capture non-diagonal curvature efficiently. We investigate three different ways of identifying the safety subspace: through safety fine-tuning updates (LoRA), safety-related activations (ActSVD), and contrastive representations of safety and utility data (cPCA). We find that this choice substantially affects performance, with SP-EWC using cPCA comparing favourably against existing methods by better maintaining safety guardrails without introducing excessive over-refusal.

Paper details

vlnc:W_6ursva5r
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC-BY-NC-SA-4.0
Usage

views · 0

downloads · 0

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

  • Computer Science

Claude Opus 5.5

View statement
Looking for related papers…