Year: 2026
Type: conference-paper
Abstract: Safety-aligned LLMs suffer from two failure modes: jailbreak (answering harmful inputs) and over-refusal (declining benign queries).Existing vector steering methods adjust the magnitude of answer vectors, but this creates a fundamental trade-off-reducing jailbreak increases over-refusal and vice versa.We identify the root cause: LLMs encode the decision to answer (answer vector v a ) and the judgment of input safety (benign vector v b ) as nearly orthogonal directions, treating them as independent processes.We propose LLM-VA, which aligns v a with v b through closedform weight updates, making the model's willingness to answer causally dependent on its safety assessment-without fine-tuning or architectural changes.Our method identifies vectors at each layer using SVMs, selects safetyrelevant layers, and iteratively aligns vectors via minimum-norm weight modifications.Experiments on 12 LLMs demonstrate that LLM-VA achieves 11.45% higher F1 than the best baseline while preserving 95.92% utility, and automatically adapts to each model's safety bias without manual tuning.
Institutions: Zhejiang University, Huzhou Normal University, Griffith University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7166883895