Another Dead End for Morphological Tags? Perturbed Inputs and Parsing
Not on the map. Why?
Abstract (arXiv)
The usefulness of part-of-speech tags for parsing has been heavily questioned due to the success of word-contextualized parsers. Yet, most studies are limited to coarse-grained tags and high quality written content; while we know little about their influence when it comes to models in production that face lexical errors. We expand these setups and design an adversarial attack to verify if the use of morphological information by parsers: (i) contributes to error propagation or (ii) if on the other hand it can play a role to correct mistakes that word-only neural parsers make. The results on 14 diverse UD treebanks show that under such attacks, for transition- and graph-based models their use contributes to degrade the performance even faster, while for the (lower-performing) sequence labeling parsers they are helpful. We also show that if morphological tags were utopically robust against lexical perturbations, they would be able to correct parsing mistakes.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 16 of its 33 references that are in the map. First posted on arXiv 24 May 2023.
Builds on
16 of 33 references.
References · 16
- 2016Enriching Word Vectors with Subword Informationcited by 1,583
- 2015Pointer Networkscited by 895
- 2018Generating Natural Language Adversarial Examplescited by 424
- 2018TextBugger: Generating Adversarial Text Against Real-world Applicationscited by 381
- 2019Combating Adversarial Misspellings with Robust Word Recognitioncited by 163
- 2017Deep Text Classification Can be Fooledcited by 106
- 2021Evaluating the Robustness of Neural Language Models to Input Perturbationscited by 87
- 2016Simple and Accurate Dependency Parsing Using Bidirectional LSTM Feature Representationscited by 55
- 2019Left-to-Right Dependency Parsing with Pointer Networkscited by 21
- 2020Bracketing Encodings for 2-Planar Dependency Parsingcited by 15
- 2015Improved Transition-Based Parsing by Modeling Characters instead of Words with LSTMscited by 10
- 2018An Investigation of the Interactions Between Pre-Trained Word Embeddings, Character Models and POS Tags in Dependency Parsingcited by 10
4 more
- 2020Is POS Tagging Necessary or Even Helpful for Neural Dependency Parsing?cited by 9
- 2021A Falta de Pan, Buenas Son Tortas: The Efficacy of Predicted UPOS Tags for Low Resource UD Parsingcited by 5
- 2021What Taggers Fail to Learn, Parsers Need the Mostcited by 2
- 2022Parsing linearizations appreciate PoS tags - but some are fussy about errorscited by 1
Built on by
Not cited yet.