SOURCE-LINKED INTELLIGENCE

Local Sparsity Enables Unsupervised LLM Safety Detection

arXiv · Artificial Intelligence · article · Sep 17, 2026 · UTC

Deployment-time safety methods for large language models (LLMs) are predominantly supervised and assume access to unsafe training data. Nevertheless, new attacks and harm categories regularly arise, not captured by models trained in such a supervised fashion. An alternative approach is to view this problem through the lens of anomaly detection, namely, to rely solely on modeling safe data and flagging out-of-distribution inputs. However, LLM activations lie in a high-dimensional space, raising concerns about whether anomaly detection is statistically feasible. We show that, under the linear re

Read original source ↗ Open in workspace

recordType

paper

region

Global