Synthetic vs. Real Reference Strings for Citation Parsing, and the Importance of Re-training and Out-Of-Sample Data for Meaningful Evaluations: Experiments with GROBID, GIANT and Cora
This addresses the data scarcity problem in citation parsing for researchers and developers, though it is incremental as it validates synthetic data rather than introducing new methods.
The study investigated whether synthetically created reference strings could effectively train citation parsing models, finding that synthetic and real data performed equally well (F1 = 0.74). It also showed that retraining improved performance by 30% and including diverse labeled fields during training boosted effectiveness by 13.5%, even if not present in evaluation data.
Citation parsing, particularly with deep neural networks, suffers from a lack of training data as available datasets typically contain only a few thousand training instances. Manually labelling citation strings is very time-consuming, hence synthetically created training data could be a solution. However, as of now, it is unknown if synthetically created reference-strings are suitable to train machine learning algorithms for citation parsing. To find out, we train Grobid, which uses Conditional Random Fields, with a) human-labelled reference strings from 'real' bibliographies and b) synthetically created reference strings from the GIANT dataset. We find that both synthetic and organic reference strings are equally suited for training Grobid (F1 = 0.74). We additionally find that retraining Grobid has a notable impact on its performance, for both synthetic and real data (+30% in F1). Having as many types of labelled fields as possible during training also improves effectiveness, even if these fields are not available in the evaluation data (+13.5% F1). We conclude that synthetic data is suitable for training (deep) citation parsing models. We further suggest that in future evaluations of reference parsers both evaluation data similar and dissimilar to the training data should be used for more meaningful evaluations.