Hierarchical Bracketing Encodings Work for Dependency Graphs
This work addresses dependency graph parsing for NLP applications, presenting an incremental improvement over existing methods.
The authors tackled dependency graph parsing by revisiting hierarchical bracketing encodings, which encode graphs as sequences for linear-time parsing while handling complex structures like reentrancies and cycles. They showed competitive results and consistent improvements in exact match accuracy on a multilingual and multi-formalism benchmark.
We revisit hierarchical bracketing encodings from a practical perspective in the context of dependency graph parsing. The approach encodes graphs as sequences, enabling linear-time parsing with $n$ tagging actions, and still representing reentrancies, cycles, and empty nodes. Compared to existing graph linearizations, this representation substantially reduces the label space while preserving structural information. We evaluate it on a multilingual and multi-formalism benchmark, showing competitive results and consistent improvements over other methods in exact match accuracy.