阿拉伯语安全对齐的选择性拒答研究:SFT、DPO 与守卫校准的实证分析
Arabic Safety Alignment as Selective Refusal: An Empirical Study of SFT, DPO, and Guard Calibration
AI 导读
一项涵盖 5 个阿拉伯语能力模型、130 次运行的实证研究显示,仅拒答目标的 SFT 会导致模型走向全面拒答,而精选混合 SFT 配置可在良性拒答率 B = 14%–23% 下达到有害提示拒答率 H = 90%–93%。
正文 · 原文
Abstract: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, while Fanar does so in two of three. Direct Preference Optimization (DPO) and inference guards change B and H differently across models rather than acting as uniform upgrades. In a blinded 300-response audit, annotator binary-refusal agreement is 89.0% (kappa = 0.78); Qwen3Guard and Aya Expanse 32B reach 88.7% and 91.0% accuracy, respectively, with no conclusive paired difference. Selected SFT raises H on Arabizi for all five models, but none reaches 90%, showing only partial transfer from Modern Standard Arabic. Overall, the results support model-specific operating-point selection: set a deployment target and retain only interventions that improve it.
| Comments: | The Fourth Arabic Natural Language Processing Conference(ArabicNLP 2026) |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2608.29378 [cs.CL] |
| (or arXiv:2608.29378v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2608.29378 arXiv-issued DOI via DataCite |
Submission history
From: Mohammad Zbib [view email]
[v1]
Sat, 29 Aug 2026 17:25:01 UTC (2,241 KB)
[v2]
Tue, 17 Nov 2026 13:26:30 UTC (2,241 KB)
来源:arXiv cs.AI · arxiv.org