{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/a-differentiable-relaxation-of-graph","title":"A Differentiable Relaxation of Graph Segmentation and Alignment for AMR Parsing","arxiv_id":"2010.12676","date":"2020-10-23","proceeding":"EMNLP 2021 11","authors":["Chunchuan Lyu","Shay B. Cohen","Ivan Titov"],"abstract":"Abstract Meaning Representations (AMR) are a broad-coverage semantic formalism which represents sentence meaning as a directed acyclic graph. To train most AMR parsers, one needs to segment the graph into subgraphs and align each such subgraph to a word in a sentence; this is normally done at preprocessing, relying on hand-crafted rules. In contrast, we treat both alignment and segmentation as latent variables in our model and induce them as part of end-to-end training. As marginalizing over the structured latent variables is infeasible, we use the variational autoencoding framework. To ensure end-to-end differentiable optimization, we introduce a differentiable relaxation of the segmentation and alignment problems. We observe that inducing segmentation yields substantial gains over using a `greedy' segmentation heuristic. The performance of our method also approaches that of a model that relies on the segmentation rules of \\citet{lyu-titov-2018-amr}, which were hand-crafted to handle individual AMR constructions.","url_abs":"https://arxiv.org/abs/2010.12676v2","url_pdf":"https://arxiv.org/pdf/2010.12676v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[{"task_slug":"amr-parsing","task_name":"AMR Parsing"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"sentence","task_name":"Sentence"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/amr-parsing-on-ldc2017t10","task":"AMR Parsing","dataset":"LDC2017T10","model":"Lyu et al. 2021. Full","rank_in_archive_order":22,"of":27,"metrics":{"Smatch":"76.1"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2010.12676","atlas_url":"https://app.syntology.ai/?focus=2010.12676","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}