nlp:abstract_meaning_representation
Table of Contents
Abstract Meaning Representation
Introductions and Overviews
- Introduction The best introduction to AMR.
-
- AMR Dict List of linguistic phenomena and how to handle them in AMR
- Generation
- Applications
Papers
See an updated list of AMR papers here: AMR Bibliography
Parsing
See also Semantic Parsing and Google Scholar - AMR Parsing.
- Graph-based
- Transition-based
- Cai & Lam 2020 - AMR Parsing via Graph Sequence Iterative Inference I would classify this approach as a transition-based algorithm that incrementally builds the graph
- Grammar-based
- Seq2seq
- AMRBART: Bai et al 2022 - Graph Pre-training for AMR Parsing and Generation (SOTA as of March 2023)
- Prompted LLMs
- Other methods
- Zhou et al 2016 - AMR Parsing with an Incremental Joint Model Has a joint model for concept and relation identification. Compares to the same feature set as JAMR.
- Domain Adaptation
Generation
See also Generation
Applications
- Machine Translation
- Summarization
- (See also Fei Liu's Publications)
- Applications to (or with) LLMs
- Question Answering
- Classification
- Information Extraction
- Dialog
- Prompting and LLMs
- Jin et al 2024 - Analyzing the Role of Semantic Representations in the Era of Large Language Models Uses AMR in a CoT-style prompt
- Fact-Checking
- Style-Transfer
- Data Augmentation
- Shou et al 2022 - AMR-DA: Data Augmentation by Abstract Meaning Representation Augments the data by parsing with AMR parser, manipulating the graph, and generating a new sentence. They used it for textual similarity task.
- Embodied or Vision and Language
- Choi et al 2022 - Scene Graph Parsing via Abstract Meaning Representation in Pre-trained Language Models Not really vision related. Uses AMR to help with caption to scene-graph parsing
- Pre-training and Embedding Representations
AMR Extensions
- Other languages than English
- Time and Temporal Information
- Quantifier Scoping and Inference
- UMR
- Interlingual extensions
- BabelNet Meaning Representation (BMR): Lorenzo et al 2022 - Fully-Semantic Parsing and Generation: the BabelNet Meaning Representation
- Dialog
Evaluation
- Parsing
- Fine-grained evaluation: Damonte et al 2016 - An Incremental Parser for Abstract Meaning Representation (see section 5)
- Generation
Software
- Parsers
- This one works well: Cai & Lam 2020 - AMR Parsing via Graph Sequence Iterative Inference Rongwen has been using it. Ask him if you want help running it.
- Libraries for reading AMR graphs
- Alignment
- Pourdamghani et al 2015 - Aligning English Strings with Abstract Meaning Representation Graphs software or this software This is the aligner that was used to produce the alignments in the AMR LDC data.
Multi-sentence AMR
Related: Coreference Resolution, Implicit Roles
- O’Gorman et al 2018 - AMR Beyond the Sentence: the Multi-sentence AMR corpus (this data is released in AMR 3.0 LDC2020T02)
- Related work: Ebner et al 2019 - Multi-Sentence Argument Linking
- Fu et al 2021 - End-to-End AMR Coreference Resolution First paper to do document coreferenece for AMRs (not full multi-document AMRs because missing implicit roles). The input is AMR, doesn't actually use the text (github)
Guesture, Situated, and Visual AMRs
- Annotation Schemes
- Parsers
Software and Resources
People
Related Pages
nlp/abstract_meaning_representation.txt · Last modified: 2025/10/17 21:56 by jmflanig