{"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/factorising-meaning-and-form-for-intent","title":"Factorising Meaning and Form for Intent-Preserving Paraphrasing","arxiv_id":"2105.15053","date":"2021-05-31","proceeding":"ACL 2021 5","authors":["Tom Hosking","Mirella Lapata"],"abstract":"We propose a method for generating paraphrases of English questions that retain the original intent but use a different surface form. Our model combines a careful choice of training objective with a principled information bottleneck, to induce a latent encoding space that disentangles meaning and form. We train an encoder-decoder model to reconstruct a question from a paraphrase with the same meaning and an exemplar with the same surface form, leading to separated encoding spaces. We use a Vector-Quantized Variational Autoencoder to represent the surface form as a set of discrete latent variables, allowing us to use a classifier to select a different surface form at test time. Crucially, our method does not require access to an external source of target exemplars. Extensive experiments and a human evaluation show that we are able to generate paraphrases with a better tradeoff between semantic preservation and syntactic novelty compared to previous methods.","url_abs":"https://arxiv.org/abs/2105.15053v1","url_pdf":"https://arxiv.org/pdf/2105.15053v1.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":[{"paper_slug":"factorising-meaning-and-form-for-intent","repo_url":"https://github.com/tomhosking/separator","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"form","task_name":"Form"},{"task_slug":"paraphrase-generation","task_name":"Paraphrase Generation"},{"task_slug":"paraphrase-identification","task_name":"Paraphrase Identification"}],"methods":[{"method_slug":"vq-vae","method_name":"VQ-VAE"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/paraphrase-generation-on-paralex","task":"Paraphrase Generation","dataset":"Paralex","model":"Separator","rank_in_archive_order":2,"of":2,"metrics":{"iBLEU":"14.84"},"uses_additional_data":false},{"leaderboard":"/sota/paraphrase-generation-on-quora-question-pairs-1","task":"Paraphrase Generation","dataset":"Quora Question Pairs","model":"Separator","rank_in_archive_order":2,"of":2,"metrics":{"iBLEU":"5.84"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2105.15053","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}