mlsanityData to 7 Oct 2026

TGIF: Tree-Graph Integrated-Format Parser for Enhanced UD with Two-Stage Generic- to Individual-Language Finetuning

arXiv 2107.06907 ↗First posted 14 Jul 2021cs.CLcited by 3

Not on the map. Why?

Abstract (arXiv)

We present our contribution to the IWPT 2021 shared task on parsing into enhanced Universal Dependencies. Our main system component is a hybrid tree-graph parser that integrates (a) predictions of spanning trees for the enhanced graphs with (b) additional graph edges not present in the spanning trees. We also adopt a finetuning strategy where we first train a language-generic parser on the concatenation of data from all available languages, and then, in a second step, finetune on each individual language separately. Additionally, we develop our own complete set of pre-processing modules relevant to the shared task, including tokenization, sentence segmentation, and multiword token expansion, based on pre-trained XLM-R models and our own pre-training of character-level language models. Our submission reaches a macro-average ELAS of 89.24 on the test set. It ranks top among all teams, with a margin of more than 2 absolute ELAS over the next best-performing submission, and best score on 16 out of 17 languages.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 15 of its 54 references that are in the map. First posted on arXiv 14 Jul 2021.

Builds on

15 of 54 references.

References · 15

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2019Unsupervised Cross-lingual Representation Learning at Scalecited by 3,907
  3. 2018Deep contextualized word representationscited by 2,285
  4. 2019On the Variance of the Adaptive Learning Rate and Beyondcited by 763
  5. 2020Stanza: A Python Natural Language Processing Toolkit for Many Human Languagescited by 719
  6. 2016Deep Biaffine Attention for Neural Dependency Parsingcited by 334
  7. 2020Universal Dependencies v2: An Evergrowing Multilingual Treebank Collectioncited by 269
  8. 2021Trankit: A Light-Weight Transformer-based Toolkit for Multilingual Natural Language Processingcited by 68
  9. 2017Simple Recurrent Units for Highly Parallelizable Recurrencecited by 60
  10. 2018Simpler but More Accurate Semantic Dependency Parsingcited by 46
  11. 2020Efficient EUD Parsingcited by 8
  12. 2020Enhanced Universal Dependency Parsing with Second-Order Inference and Mixture of Training Datacited by 5
3 more
  1. 2020Kopsala: Transition-Based Graph Parsing via Efficient Training and Effective Encodingcited by 4
  2. 2021Coordinate Constructions in English Enhanced Universal Dependencies: Analysis and Computational Modelingcited by 4
  3. 2020The ADAPT Enhanced Dependency Parser at the IWPT 2020 Shared Taskcited by 3

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 2.

3 papers.

  1. 2022DAMO-NLP at SemEval-2022 Task 11: A Knowledge-based System for Multilingual Named Entity Recognitioncited by 17
  2. 2023DAMO-NLP at SemEval-2023 Task 2: A Unified Retrieval-augmented System for Multilingual Named Entity Recognitioncited by 9
  3. 2023A Semi-Autoregressive Graph Generative Model for Dependency Graph Parsingcited by 0