mlsanityData to 14 Oct 2026

Semantic Role Labeling as Dependency Parsing: Exploring Latent Tree Structures Inside Arguments

arXiv 2110.06865 ↗First posted 13 Oct 2021cs.CLcited by 14

Not on the map. Why?

Abstract (arXiv)

Semantic role labeling (SRL) is a fundamental yet challenging task in the NLP community. Recent works of SRL mainly fall into two lines: 1) BIO-based; 2) span-based. Despite ubiquity, they share some intrinsic drawbacks of not considering internal argument structures, potentially hindering the model's expressiveness. The key challenge is arguments are flat structures, and there are no determined subtree realizations for words inside arguments. To remedy this, in this paper, we propose to regard flat argument spans as latent subtrees, accordingly reducing SRL to a tree parsing task. In particular, we equip our formulation with a novel span-constrained TreeCRF to make tree structures span-aware and further extend it to the second-order case. We conduct extensive experiments on CoNLL05 and CoNLL12 benchmarks. Results reveal that our methods perform favorably better than all previous syntax-agnostic works, achieving new state-of-the-art under both end-to-end and w/ gold predicates settings.

Key facts. Cited by 14 papers in the map (data through 14 Oct 2026). Builds on 50 of its 126 references that are in the map. First posted on arXiv 13 Oct 2021.

Builds on

50 of 126 references.

References · 50

  1. 2017Attention Is All You Needcited by 60,361
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  3. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  4. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
  5. 2019XLNet: Generalized Autoregressive Pretraining for Language Understandingcited by 2,811
  6. 2016Neural Architectures for Named Entity Recognitioncited by 550
  7. 2016Deep Biaffine Attention for Neural Dependency Parsingcited by 334
  8. 2021Structured Prediction as Translation between Augmented Natural Languagescited by 183
  9. 2019Simple BERT Models for Relation Extraction and Semantic Role Labelingcited by 154
  10. 2020Named Entity Recognition as Dependency Parsingcited by 151
  11. 2017Encoding Sentences with Graph Convolutional Networks for Semantic Role Labelingcited by 137
  12. 2017Structured Attention Networkscited by 95
38 more
  1. 2018Linguistically-Informed Self-Attention for Semantic Role Labelingcited by 71
  2. 2019AMR Parsing as Sequence-to-Graph Transductioncited by 50
  3. 2020Efficient Second-Order TreeCRF for Neural Dependency Parsingcited by 48
  4. 2020Torch-Struct: Deep Structured Prediction Librarycited by 40
  5. 2018Jointly Predicting Predicates and Arguments in Neural Semantic Role Labelingcited by 34
  6. 2020StructFormer: Joint Unsupervised Induction of Dependency and Constituency Structure from Masked Language Modelingcited by 33
  7. 2019Broad-Coverage Semantic Parsing as Transductioncited by 32
  8. 2021The Stem Cell Hypothesis: Dilemma behind Multi-Task Learning with Transformer Encoderscited by 32
  9. 2019Dependency-Guided LSTM-CRF for Named Entity Recognitioncited by 28
  10. 2018Syntactic Scaffolds for Semantic Structurescited by 27
  11. 2017Learning Structured Text Representationscited by 24
  12. 2017Deep Semantic Role Labeling with Self-Attentioncited by 23
  13. 2020Nested Named Entity Recognition with Partially-Observed TreeCRFscited by 23
  14. 2019Dependency or Span, End-to-End Uniform Semantic Role Labelingcited by 22
  15. 2021Better Feature Integration for Named Entity Recognitioncited by 22
  16. 2019Second-Order Semantic Dependency Parsing with End-to-End Neural Networkscited by 21
  17. 2020The Return of Lexical Dependencies: Neural Lexicalized PCFGscited by 19
  18. 2021Neural Bi-Lexicalized PCFG Inductioncited by 18
  19. 2018A Span Selection Model for Semantic Role Labelingcited by 15
  20. 2018A Full End-to-End Semantic Role Labeler, Syntax-agnostic Over Syntax-aware?cited by 13
  21. 2019Parsing All: Syntax and Semantics, Dependencies and Spanscited by 13
  22. 2021Headed-Span-Based Projective Dependency Parsingcited by 11
  23. 2020Structured Tuning for Semantic Role Labelingcited by 10
  24. 2020Second-Order Unsupervised Neural Dependency Parsingcited by 10
  25. 2019Syntax-aware Neural Semantic Role Labelingcited by 8
  26. 2021Syntax-Aware Graph-to-Graph Transformer for Semantic Role Labellingcited by 8
  27. 2017Dependency Grammar Induction with Neural Lexicalization and Big Training Datacited by 7
  28. 2020Syntax Role for Neural Semantic Role Labelingcited by 7
  29. 2017CRF Autoencoder for Unsupervised Dependency Parsingcited by 6
  30. 2020Improved Semantic Role Labeling using Parameterized Neighborhood Memory Adaptationcited by 6
  31. 2021Combining (second-order) graph-based and headed-span-based projective dependency parsingcited by 5
  32. 2015Unsupervised Dependency Parsing: Let's Use Supervised Parserscited by 3
  33. 2019HLT@SUDA at SemEval 2019 Task 1: UCCA Graph Parsing as Constituent Tree Parsingcited by 3
  34. 2020Semantic Role Labeling as Syntactic Dependency Parsingcited by 3
  35. 2015Probabilistic Models for High-Order Projective Dependency Parsingcited by 1
  36. 2016Latent Dependency Forest Modelscited by 1
  37. 2020Neural Latent Dependency Model for Sequence Labelingcited by 1
  38. 2021Splitting EUD graphs into trees: A quick and clatty approachcited by 1

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 4; 2023: 5; 2024: 3; 2025: 1.

14 papers.

Sort:

  1. 2023Pushing the Limits of ChatGPT on NLP Taskscited by 27
  2. 2022Nested Named Entity Recognition as Latent Lexicalized Constituency Parsingcited by 14
  3. 2024Potential and Limitations of LLMs in Capturing Structured Semantics: A Case Study on SRLcited by 10
  4. 2023Improving Aspect-Based Sentiment with End-to-End Semantic Role Labeling Modelcited by 5
  5. 2023Efficient Beam Tree Recursioncited by 4
  6. 2022Do Transformers Encode a Foundational Ontology? Probing Abstract Classes in Natural Languagecited by 3
  7. 2022Transition-based Semantic Role Labeling with Pointer Networkscited by 3
  8. 2021Fast and Accurate End-to-End Span-based Semantic Role Labeling as Word-based Graph Parsingcited by 1
  9. 2023Structured Sentiment Analysis as Transition-based Dependency Graph Parsingcited by 1
  10. 2024Revisiting Structured Sentiment Analysis as Latent Dependency Graph Parsingcited by 1
  11. 2025LLMs Can Also Do Well! Breaking Barriers in Semantic Role Labeling via Large Language Modelscited by 1
  12. 2022PriMeSRL-Eval: A Practical Quality Metric for Semantic Role Labeling Systems Evaluationcited by 0
2 more
  1. 2023EDeR: A Dataset for Exploring Dependency Relations Between Eventscited by 0
  2. 2024Character-Level Chinese Dependency Parsing via Modeling Latent Intra-Word Structurecited by 0