AI-generated text is nowadays produced at scale across domains and heterogeneous generation pipelines, making robustness to distribution shift a central requirement for supervised binary detectors. We train transformer-based detectors on HC3 PLUS and calibrate a single decision threshold by maximising balanced accuracy on held-out validation; this threshold is then kept fixed for all downstream test distributions, revealing domain- and generator-dependent error asymmetries under shift. We evaluate in-domain on HC3 PLUS, under cross-dataset transfer to the multi-domain, multi-generator M4 benchmark, and on the external AI-Text-Detection-Pile. Although base models achieve near-ceiling in-domain performance (up to 99.5% balanced accuracy), performance under shift is brittle and strongly model-dependent. Feature augmentation via attention-based linguistic feature fusion improves transfer, with our best model (DeBERTa-v3-base+FeatAttn) achieving 85.9% balanced accuracy on M4. Multi-seed experiments confirm high stability. Under the same fixed-threshold protocol, our model outperforms strong zero-shot baselines by up to +7.22 points. Category-level ablations further show that readability and vocabulary features contribute most to robustness under shift. Overall, these results demonstrate that feature augmentation and a modern DeBERTa backbone significantly outperform earlier BERT/RoBERTa models, while the fixed-threshold protocol provides a more realistic and informative assessment of practical detector robustness.

翻译：如今，AI生成的文本在跨领域和异构生成流水线中大规模产生，使得对分布偏移的鲁棒性成为有监督二分类器的核心需求。我们在HC3 PLUS数据集上训练基于Transformer的检测器，并通过最大化留出验证集上的平衡准确率来校准单一决策阈值；该阈值随后对所有下游测试分布保持固定，揭示了偏移下领域和生成器相关的错误不对称性。我们在HC3 PLUS上进行域内评估，在跨数据集迁移至多领域、多生成器的M4基准测试中，以及外部AI-Text-Detection-Pile数据集上。尽管基础模型在域内表现接近上限（平衡准确率高达99.5%），但偏移下的性能脆弱且高度依赖模型。通过基于注意力的语言特征融合进行特征增强改善了迁移性能，我们的最佳模型（DeBERTa-v3-base+FeatAttn）在M4上达到85.9%的平衡准确率。多随机种子实验证实了高稳定性。在同一固定阈值协议下，我们的模型比强零样本基线高出最多7.22个百分点。类别级消融实验进一步表明，可读性和词汇特征对偏移下的鲁棒性贡献最大。总体而言，这些结果证明特征增强与现代DeBERTa骨干网络显著优于早期BERT/RoBERTa模型，而固定阈值协议则提供了对实用检测器鲁棒性更现实且更具信息量的评估。