Making Your LLMs More Objective: Stabilizing LLM Safety Behavior Across Traits with Trait-Invariant Safety Tuning

Aligned large language models (LLMs) are expected to exhibit safety behavior based on the content of the user request: they should refuse unsafe requests and comply with safe ones. However, we show that the same request can elicit substantially different safety decisions under different traits assigned in the system prompt, a failure mode we call trait-induced safety variation. To measure this failure, we introduce refusal-based metrics: Trait-Induced Deviation measures dataset-level deviation f

Related evidence

These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.

No strong metadata relationship is available in the current record.

How to cite this record

ethics.ai (12 August 2026), “Making Your LLMs More Objective: Stabilizing LLM Safety Behavior Across Traits with Trait-Invariant Safety Tuning,” evidence record 18791, https://ethics.ai/record/18791 (originally published by arXiv).

Use and limitations

This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.