Xia Cui, Danushka Bollegala. Proceedings of the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing. 2020.
Selecting pivot features that connect a source domain to a target domain is an important first step in unsupervised domain adaptation (UDA). Although different strategies such as the frequency of a feature in a domain, mutual (or pointwise mutual) information have been proposed in prior work in domain adaptation (DA) for selecting pivots, a comparative study into (a) how the pivots selected using existing strategies differ, and (b) how the pivot selection strategy affects the performance of a target DA task remain unknown. In this paper, we perform a comparative study covering different strategies that use both labelled (available for the source domain only) as well as unlabelled (available for both the source and target domains) data for selecting pivots for UDA. Our experiments show that in most cases pivot selection strategies that use labelled data outperform their unlabelled counterparts, emphasising the importance of the source domain labelled data for UDA. Moreover, pointwise mutual information and frequency-based pivot selection strategies obtain the best performances in two state-of-the-art UDA methods.
Domain adaptation considers the problem of generalising a model learnt using data from a particular source domain to a different target domain. Often it is difficult to find a suitable single source to adapt from, and one must consider multiple sources. Using an unrelated source can result in sub-optimal performance, known as the \emph{negative transfer}. However, it is challenging to select the appropriate source(s) for classifying a given target instance in multi-source unsupervised domain adaptation (UDA). We model source-selection as an attention-learning problem, where we learn attention over sources for a given target instance. For this purpose, we first independently learn source-specific classification models, and a relatedness map between sources and target domains using pseudo-labelled target domain instances. Next, we learn attention-weights over the sources for aggregating the predictions of the source-specific models. Experimental results on cross-domain sentiment classification benchmarks show that the proposed method outperforms prior proposals in multi-source UDA.
Subjective text classification tasks, such as abuse detection and stance analysis, often suffer from high levels of annotator disagreement. Conventional approaches typically collapse these disagreements into a single ground truth, thereby discarding valuable supervision signals. We propose MO-WEL (Multi-Objective Weighted Ensemble Learning), a novel framework that explicitly leverages annotator disagreement by jointly optimising ensemble weights and size under multiple objectives. Candidate predictors are trained on diverse label projections obtained through random sampling or annotator-specific selection, and ensemble weights are optimised with respect to three complementary losses: F1 score, cross-entropy and Manhattan distance, alongside a regularisation term. Experiments on four benchmark datasets (ConvAbuse, HS-Brexit, MD-Agreement and ArMIS) show that MO-WEL consistently outperforms strong baselines in accuracy, calibration, and distributional alignment. A case study further demonstrates that MO-WEL produces predictions that balance majority correctness with minority annotator perspectives, yielding interpretable and reliable outputs. Our findings highlight the importance of modelling annotator diversity and suggest ensemble optimisation as a principled means of incorporating disagreement into subjective NLP tasks.