SOURCE-LINKED INTELLIGENCE

Interpreting and Steering for Safe and Correct Code Generation

arXiv · AI, language, vision and robotics · article · Aug 30, 2026 · UTC

Large language models (LLMs) frequently generate source code containing vulnerabilities, yet little work studies the internal mechanisms that distinguish safe from vulnerable generation in them. In this work, we systematically perform a mechanistic interpretation of LLMs, aiming at both understanding how code safety-vs-vulnerability is represented or driven by components in an LM and turning the insights into actionable steering strategies to encourage safer code generation. To this end, we introduce CodeSec-Pairs, a dataset of 9,342 Python safe-and-vulnerable contrastive code pairs, sampled f

Read original source ↗ Open in workspace

recordType

paper

region

Global