Normalizing Compositional Structures Across Graphbanks
This work addresses challenges in uniform linguistic analysis and cross-framework semantic parsing for researchers and practitioners in computational linguistics, though it is incremental as it builds on existing normalization techniques.
The paper tackled the problem of structural differences between graph-based meaning representations (MRs) by developing a methodology to normalize compositional discrepancies, finding that most divergent phenomena can be normalized using linguistically-grounded rules, which improved multi-task learning in low-resource settings.
The emergence of a variety of graph-based meaning representations (MRs) has sparked an important conversation about how to adequately represent semantic structure. These MRs exhibit structural differences that reflect different theoretical and design considerations, presenting challenges to uniform linguistic analysis and cross-framework semantic parsing. Here, we ask the question of which design differences between MRs are meaningful and semantically-rooted, and which are superficial. We present a methodology for normalizing discrepancies between MRs at the compositional level (Lindemann et al., 2019), finding that we can normalize the majority of divergent phenomena using linguistically-grounded rules. Our work significantly increases the match in compositional structure between MRs and improves multi-task learning (MTL) in a low-resource setting, demonstrating the usefulness of careful MR design analysis and comparison.