用课程学习和伪标签提升阿拉伯方言识别模型泛化能力
Curriculum Learning and Pseudo-Labeling Improve the Generalization of Multi-Label Arabic Dialect Identification Models
- 基于大模型生成多标签数据，解决标注难问题
- 引入课程学习策略，按方言复杂度逐步训练，提升性能
- 在公开榜单上达到0.69宏F1，优于此前最佳系统
长期以来，阿拉伯方言识别（ADI）被当作单标签分类任务处理。近期研究认为应改为多标签分类。然而，现有资源受限于单标签数据集，缺乏大规模多标签训练数据。通过分析单标签训练模型，我们发现将此类数据用于多标签任务的主要难点在于负样本选择——许多被标记为负的句子实际上可能属于多个方言。为此，我们利用GPT-4o与二值方言可接受性分类器生成自动多标签标注，并以阿拉伯方言程度（ALDi）为指导进行聚合，构建多标签数据集。随后，采用与方言复杂度和标签基数对齐的课程学习策略，训练基于BERT的多标签分类器。在MLADI排行榜上，最优模型LAHJATBERT取得0.69的宏F1，优于此前最强系统（0.55）。代码与数据已公开。
原文摘要 · Abstract (English)
Being modeled as a single-label classification task for a long time, recent work has argued that Arabic Dialect Identification (ADI) should be framed as a multi-label classification task. However, ADI remains constrained by the availability of single-label datasets, with no large-scale multi-label resources available for training. By analyzing models trained on single-label ADI data, we show that the main difficulty in repurposing such datasets for Multi-Label Arabic Dialect Identification (MLADI) lies in the selection of negative samples, as many sentences treated as negative could be acceptable in multiple dialects. To address these issues, we construct a multi-label dataset by generating automatic multi-label annotations using GPT-4o and binary dialect acceptability classifiers, with aggregation guided by the Arabic Level of Dialectness (ALDi). Afterward, we train a BERT-based multi-label classifier using curriculum learning strategies aligned with dialectal complexity and label cardinality. On the MLADI leaderboard, our best-performing LAHJATBERT model achieves a macro F1 of 0.69, compared to 0.55 for the strongest previously reported system. Code and data are available at https://mohamedalaa9.github.io/lahjatbert/.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。