SOURCE-LINKED INTELLIGENCE

Arabic Safety Alignment as Selective Refusal: An Empirical Study of SFT, DPO, and Guard Calibration

arXiv · AI, language, vision and robotics · article · Aug 29, 2026 · UTC

Arabic large language models must refuse harmful prompts without over-refusing benign or sensitive prompts, yet a single refusal rate hides this trade-off. We evaluate it using benign refusal B and harmful-prompt refusal H, where H measures refusal rather than harmful compliance. Across five Arabic-capable models and 130 runs on the full human-written AraSafe set, refusal-only supervised fine-tuning (SFT) collapses toward blanket refusal, whereas selected mixed-SFT configurations reach H = 90% to 93% at B = 14% to 23%; four selected configurations exceed the H = 90% target in all three runs, w

Read original source ↗ Open in workspace

recordType

paper

region

Global