mlsanityData to 7 Oct 2026

Inducing and Using Alignments for Transition-based AMR Parsing

arXiv 2205.01464 ↗First posted 3 May 2022cs.CLcited by 11

Not on the map. Why?

Abstract (arXiv)

Transition-based parsers for Abstract Meaning Representation (AMR) rely on node-to-word alignments. These alignments are learned separately from parser training and require a complex pipeline of rule-based components, pre-processing, and post-processing to satisfy domain-specific constraints. Parsers also train on a point-estimate of the alignment pipeline, neglecting the uncertainty due to the inherent ambiguity of alignment. In this work we explore two avenues for overcoming these limitations. First, we propose a neural aligner for AMR that learns node-to-word alignments without relying on complex pipelines. We subsequently explore a tighter integration of aligner and parser training by considering a distribution over oracle action sequences arising from aligner uncertainty. Empirical results show this approach leads to more accurate alignments and generalization better from the AMR2.0 to AMR3.0 corpora. We attain a new state-of-the art for gold-only trained models, matching silver-trained performance without the need for beam search on AMR3.0.

Key facts. Cited by 11 papers in the map (data through 7 Oct 2026). Builds on 34 of its 50 references that are in the map. First posted on arXiv 3 May 2022.

Builds on

34 of 50 references.

References · 34

  1. 2019BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehensioncited by 5,874
  2. 2018Deep contextualized word representationscited by 2,285
  3. 2015Show, Attend and Tell: Neural Image Caption Generation with Visual Attentioncited by 1,421
  4. 2015Effective Approaches to Attention-based Neural Machine Translationcited by 1,024
  5. 2015Importance Weighted Autoencoderscited by 454
  6. 2017Online and Linear-Time Attention by Enforcing Monotonic Alignmentscited by 85
  7. 2016Variational inference for Monte Carlo objectivescited by 83
  8. 2022Graph Pre-training for AMR Parsing and Generationcited by 55
  9. 2019AMR Parsing as Sequence-to-Graph Transductioncited by 50
  10. 2020Transition-based Parsing with Stack-Transformerscited by 37
  11. 2018AMR Parsing as Graph Prediction with Latent Alignmentcited by 36
  12. 2018Latent Alignment and Variational Attentioncited by 36
22 more
  1. 2019Broad-Coverage Semantic Parsing as Transductioncited by 32
  2. 2016Neural Machine Translation with Supervised Attentioncited by 31
  3. 2020End-to-End Neural Word Alignment Outperforms GIZA++cited by 28
  4. 2021Maximum Bayes Smatch Ensemble Distillation for AMR Parsingcited by 27
  5. 2019Rewarding Smatch: Transition-Based AMR Parsing with Reinforcement Learningcited by 25
  6. 2021AMR Parsing with Action-Pointer Transformercited by 25
  7. 2021Structure-aware Fine-tuning of Sequence-to-sequence Transformers for Transition-based AMR Parsingcited by 25
  8. 2019Exact Hard Monotonic Attention for Character-Level Transductioncited by 22
  9. 2020Mask-Align: Self-Supervised Neural Word Alignmentcited by 20
  10. 2016Incorporating Structural Alignment Biases into an Attentional Neural Translation Modelcited by 16
  11. 2020Posterior Control of Blackbox Generationcited by 16
  12. 2020Pushing the Limits of AMR Parsing with Self-Learningcited by 16
  13. 2017AMR Parsing using Stack-LSTMscited by 15
  14. 2016Online Segment to Segment Neural Transductioncited by 14
  15. 2018Hard Non-Monotonic Attention for Character-Level Transductioncited by 12
  16. 2018An AMR Aligner Tuned by Transition-based Parsercited by 12
  17. 2015Learning Wake-Sleep Recurrent Attention Modelscited by 9
  18. 2021Probabilistic, Structure-Aware Algorithms for Improved Variety, Accuracy, and Coverage of AMR Alignmentscited by 9
  19. 2020A Differentiable Relaxation of Graph Segmentation and Alignment for AMR Parsingcited by 7
  20. 2016Training with Exploration Improves a Greedy Stack-LSTM Parsercited by 4
  21. 2018Dependency Grammar Induction with a Neural Variational Transition-based Parsercited by 4
  22. 2020Neural Baselines for Word Alignmentcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 2; 2023: 2; 2024: 4; 2025: 2; 2026: 1.

11 papers.

  1. 2023Effective Distillation of Table-based Reasoning Ability from LLMscited by 23
  2. 2023Neuro-symbolic Commonsense Social Reasoningcited by 5
  3. 2022Visual Semantic Parsing: From Images to Abstract Meaning Representationcited by 4
  4. 2024MASSIVE Multilingual Abstract Meaning Representation: A Dataset and Baselines for Hallucination Detectioncited by 4
  5. 2024Understanding Enthymemes in Argument Maps: Bridging Argument Mining and Logic-based Argumentationcited by 2
  6. 2025Can LLMs Interpret and Leverage Structured Linguistic Representations? A Case Study with AMRscited by 2
  7. 2024Rematch: Robust and Efficient Matching of Local Knowledge Graphs to Improve Structural and Semantic Similaritycited by 1
  8. 2022A Two-Stage Method for Chinese AMR Parsingcited by 0
  9. 2024FLAG: Financial Long Document Classification via AMR-based GNNcited by 0
  10. 2025EventRR: Event Referential Reasoning for Referring Video Object Segmentationcited by 0
  11. 2026On the (In)effectiveness of AMR Augmentation for Large Language Modelscited by 0