# Inducing and Using Alignments for Transition-based AMR Parsing

arXiv: https://arxiv.org/abs/2205.01464 · HTML: https://mlsanity.com/paper/2205.01464/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 50 references that are in the map.
- First posted on arXiv 3 May 2022.

## Builds on (most cited first, 30 of 34)

- [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://mlsanity.com/paper/1910.13461/) (2019)
- [Deep contextualized word representations](https://mlsanity.com/paper/1802.05365/) (2018)
- [Show, Attend and Tell: Neural Image Caption Generation with Visual Attention](https://mlsanity.com/paper/1502.03044/) (2015)
- [Effective Approaches to Attention-based Neural Machine Translation](https://mlsanity.com/paper/1508.04025/) (2015)
- [Importance Weighted Autoencoders](https://mlsanity.com/paper/1509.00519/) (2015)
- [Online and Linear-Time Attention by Enforcing Monotonic Alignments](https://mlsanity.com/paper/1704.00784/) (2017)
- [Variational inference for Monte Carlo objectives](https://mlsanity.com/paper/1602.06725/) (2016)
- [Graph Pre-training for AMR Parsing and Generation](https://mlsanity.com/paper/2203.07836/) (2022)
- [AMR Parsing as Sequence-to-Graph Transduction](https://mlsanity.com/paper/1905.08704/) (2019)
- [Transition-based Parsing with Stack-Transformers](https://mlsanity.com/paper/2010.10669/) (2020)
- [AMR Parsing as Graph Prediction with Latent Alignment](https://mlsanity.com/paper/1805.05286/) (2018)
- [Latent Alignment and Variational Attention](https://mlsanity.com/paper/1807.03756/) (2018)
- [Broad-Coverage Semantic Parsing as Transduction](https://mlsanity.com/paper/1909.02607/) (2019)
- [Neural Machine Translation with Supervised Attention](https://mlsanity.com/paper/1609.04186/) (2016)
- [Maximum Bayes Smatch Ensemble Distillation for AMR Parsing](https://mlsanity.com/paper/2112.07790/) (2021)
- [End-to-End Neural Word Alignment Outperforms GIZA++](https://mlsanity.com/paper/2004.14675/) (2020)
- [Rewarding Smatch: Transition-Based AMR Parsing with Reinforcement Learning](https://mlsanity.com/paper/1905.13370/) (2019)
- [AMR Parsing with Action-Pointer Transformer](https://mlsanity.com/paper/2104.14674/) (2021)
- [Structure-aware Fine-tuning of Sequence-to-sequence Transformers for Transition-based AMR Parsing](https://mlsanity.com/paper/2110.15534/) (2021)
- [Exact Hard Monotonic Attention for Character-Level Transduction](https://mlsanity.com/paper/1905.06319/) (2019)
- [Mask-Align: Self-Supervised Neural Word Alignment](https://mlsanity.com/paper/2012.07162/) (2020)
- [Incorporating Structural Alignment Biases into an Attentional Neural Translation Model](https://mlsanity.com/paper/1601.01085/) (2016)
- [Posterior Control of Blackbox Generation](https://mlsanity.com/paper/2005.04560/) (2020)
- [Pushing the Limits of AMR Parsing with Self-Learning](https://mlsanity.com/paper/2010.10673/) (2020)
- [AMR Parsing using Stack-LSTMs](https://mlsanity.com/paper/1707.07755/) (2017)
- [Online Segment to Segment Neural Transduction](https://mlsanity.com/paper/1609.08194/) (2016)
- [Hard Non-Monotonic Attention for Character-Level Transduction](https://mlsanity.com/paper/1808.10024/) (2018)
- [An AMR Aligner Tuned by Transition-based Parser](https://mlsanity.com/paper/1810.03541/) (2018)
- [Learning Wake-Sleep Recurrent Attention Models](https://mlsanity.com/paper/1509.06812/) (2015)
- [Probabilistic, Structure-Aware Algorithms for Improved Variety, Accuracy, and Coverage of AMR Alignments](https://mlsanity.com/paper/2106.06002/) (2021)

## Built on by (most cited first, 12 of 12)

- [Effective Distillation of Table-based Reasoning Ability from LLMs](https://mlsanity.com/paper/2309.13182/) (2023)
- [Neuro-symbolic Commonsense Social Reasoning](https://mlsanity.com/paper/2303.08264/) (2023)
- [Visual Semantic Parsing: From Images to Abstract Meaning Representation](https://mlsanity.com/paper/2210.14862/) (2022)
- [MASSIVE Multilingual Abstract Meaning Representation: A Dataset and Baselines for Hallucination Detection](https://mlsanity.com/paper/2405.19285/) (2024)
- [Can LLMs Interpret and Leverage Structured Linguistic Representations? A Case Study with AMRs](https://mlsanity.com/paper/2504.04745/) (2025)
- [Understanding Enthymemes in Argument Maps: Bridging Argument Mining and Logic-based Argumentation](https://mlsanity.com/paper/2408.08648/) (2024)
- [Rematch: Robust and Efficient Matching of Local Knowledge Graphs to Improve Structural and Semantic Similarity](https://mlsanity.com/paper/2404.02126/) (2024)
- [A Two-Stage Method for Chinese AMR Parsing](https://mlsanity.com/paper/2209.14512/) (2022)
- [FLAG: Financial Long Document Classification via AMR-based GNN](https://mlsanity.com/paper/2410.02024/) (2024)
- [EventRR: Event Referential Reasoning for Referring Video Object Segmentation](https://mlsanity.com/paper/2508.07171/) (2025)
- [On the (In)effectiveness of AMR Augmentation for Large Language Models](https://mlsanity.com/paper/2609.40121/) (2026)
- [How Reliable are AMR Parser Outputs? Uncertainty Estimation for AMR Parsing](https://mlsanity.com/paper/2610.13236/) (2026)

## 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.