Evaluating Morphological Plausibility of Subword Tokenization via Statistical Alignment with Morpho-Syntactic Features
This provides a more accessible evaluation method for subword tokenization across diverse languages, though it is incremental as it builds on existing alignment techniques.
The paper tackles the problem of evaluating morphological plausibility in subword tokenization by introducing a novel metric that uses morpho-syntactic features, showing it correlates well with traditional morpheme boundary recall and is more broadly applicable across languages.
We present a novel metric for the evaluation of the morphological plausibility of subword segmentation. Unlike the typically used morpheme boundary or retrieval F-score, which requires gold segmentation data that is either unavailable or of inconsistent quality across many languages, our approach utilizes morpho-syntactic features. These are available in resources such as Universal Dependencies or UniMorph for a much wider range of languages. The metric works by probabilistically aligning subwords with morphological features through an IBM Model 1. Our experiments show that the metric correlates well with traditional morpheme boundary recall while being more broadly applicable across languages with different morphological systems.