Pretrained language models (PLMs) are trained on massive corpora, but often need to specialize to specific domains. A parameter-efficient adaptation method suggests training an adapter for each domain on the task of language modeling. This leads to good in-domain scores but can be impractical for domain- or resource-restricted settings. A solution is to use a related-domain adapter for the novel domain at test time. In this paper, we introduce AdapterSoup, an approach that performs weight-space averaging of adapters trained on different domains. Our approach is embarrassingly parallel: first, we train a set of domain-specific adapters; then, for each novel domain, we determine which adapters should be averaged at test time. We present extensive experiments showing that AdapterSoup consistently improves performance to new domains without extra training. We also explore weight averaging of adapters trained on the same domain with different hyper-parameters, and show that it preserves the performance of a PLM on new domains while obtaining strong in-domain results. We explore various approaches for choosing which adapters to combine, such as text clustering and semantic similarity. We find that using clustering leads to the most competitive results on novel domains.

翻译：预训练语言模型在大型语料库上训练，但通常需要针对特定领域进行专门化。一种参数高效的适应方法建议在语言建模任务上为每个领域训练一个适配器。这虽然能带来良好的领域内评分，但在领域受限或资源受限的场景下可能不切实际。一种解决方案是在测试阶段对新颖领域使用相关领域的适配器。本文提出了适配器汤（AdapterSoup）方法，该方法对不同领域训练的适配器进行权重空间平均。我们的方法是令人尴尬的并行化：首先，我们训练一组领域特异性适配器；然后，对于每个新颖领域，我们确定在测试阶段应该平均哪些适配器。我们进行的广泛实验表明，适配器汤无需额外训练即可持续提升在新领域的性能。我们还探索了对同一领域使用不同超参数训练的适配器进行权重平均，并证明它能在保持预训练语言模型在新领域性能的同时，获得强大的领域内结果。我们研究了多种选择适配器组合的方法，例如文本聚类和语义相似度。我们发现，使用聚类能为新颖领域带来最具竞争力的结果。