Aspect-based Sentiment Analysis (ABSA) aims to determine the sentiment polarity towards an aspect. Because of the expensive and limited labelled data, the pretraining strategy has become the de-facto standard for ABSA. However, there always exists severe domain shift between the pretraining and downstream ABSA datasets, hindering the effective knowledge transfer when directly finetuning and making the downstream task performs sub-optimal. To mitigate such domain shift, we introduce a unified alignment pretraining framework into the vanilla pretrain-finetune pipeline with both instance- and knowledge-level alignments. Specifically, we first devise a novel coarse-to-fine retrieval sampling approach to select target domain-related instances from the large-scale pretraining dataset, thus aligning the instances between pretraining and target domains (First Stage). Then, we introduce a knowledge guidance-based strategy to further bridge the domain gap at the knowledge level. In practice, we formulate the model pretrained on the sampled instances into a knowledge guidance model and a learner model, respectively. On the target dataset, we design an on-the-fly teacher-student joint fine-tuning approach to progressively transfer the knowledge from the knowledge guidance model to the learner model (Second Stage). Thereby, the learner model can maintain more domain-invariant knowledge when learning new knowledge from the target dataset. In the Third Stage, the learner model is finetuned to better adapt its learned knowledge to the target dataset. Extensive experiments and analyses on several ABSA benchmarks demonstrate the effectiveness and universality of our proposed pretraining framework. Our source code and models are publicly available at https://github.com/WHU-ZQH/UIKA.

翻译：面向方面情感分析（ABSA）旨在判断给定方面的情感极性。由于标注数据昂贵且有限，预训练策略已成为ABSA的事实标准。然而，预训练数据集与下游ABSA数据集之间始终存在严重的领域偏移，这阻碍了直接微调时的有效知识迁移，并导致下游任务性能次优。为了缓解这种领域偏移，我们提出了一种统一的实例级与知识级对齐预训练框架，将其融入传统的预训练-微调流程中。具体而言，我们首先设计了一种新颖的由粗到细的检索采样方法，从大规模预训练数据集中选取与目标领域相关的实例，从而实现预训练领域与目标领域之间的实例对齐（第一阶段）。随后，我们引入了一种基于知识引导的策略，进一步在知识层面弥合领域差距。实践中，我们将基于采样实例预训练的模型分别构建为知识引导模型和学习器模型。在目标数据集上，我们设计了一种动态师生联合微调方法，逐步将知识从知识引导模型迁移至学习器模型（第二阶段）。由此，学习器模型在从目标数据集学习新知识时，能够保留更多领域不变知识。在第三阶段，我们对学习器模型进行微调，使其学得的知识更好地适应目标数据集。在多个ABSA基准上的大量实验与分析表明，我们提出的预训练框架具有有效性和通用性。我们的源代码和模型已公开发布于https://github.com/WHU-ZQH/UIKA。