{"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/discovering-the-compositional-structure-of","title":"Discovering the Compositional Structure of Vector Representations with Role Learning Networks","arxiv_id":"1910.09113","date":"2019-10-21","proceeding":"EMNLP (BlackboxNLP) 2020 11","authors":["Paul Soulos","Tom McCoy","Tal Linzen","Paul Smolensky"],"abstract":"How can neural networks perform so well on compositional tasks even though they lack explicit compositional representations? We use a novel analysis technique called ROLE to show that recurrent neural networks perform well on such tasks by converging to solutions which implicitly represent symbolic structure. This method uncovers a symbolic structure which, when properly embedded in vector space, closely approximates the encodings of a standard seq2seq network trained to perform the compositional SCAN task. We verify the causal importance of the discovered symbolic structure by showing that, when we systematically manipulate hidden embeddings based on this symbolic structure, the model's output is changed in the way predicted by our analysis.","url_abs":"https://arxiv.org/abs/1910.09113v3","url_pdf":"https://arxiv.org/pdf/1910.09113v3.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":"discovering-the-compositional-structure-of","repo_url":"https://github.com/iclr2020-anonymous1/role-learner","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"discovering-the-compositional-structure-of","repo_url":"https://github.com/psoulos/role-decomposition","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"sentence-embedding","task_name":"Sentence Embedding"}],"methods":[{"method_slug":"gru","method_name":"GRU"},{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"seq2seq","method_name":"Seq2Seq"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1910.09113","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}