{"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/scalable-knowledge-base-completion-with","title":"Scalable knowledge base completion with superposition memories","arxiv_id":"2110.12341","date":"2021-10-24","proceeding":null,"authors":["Matthias Lalisse","Eric Rosen","Paul Smolensky"],"abstract":"We present Harmonic Memory Networks (HMem), a neural architecture for knowledge base completion that models entities as weighted sums of pairwise bindings between an entity's neighbors and corresponding relations. Since entities are modeled as aggregated neighborhoods, representations of unseen entities can be generated on the fly. We demonstrate this with two new datasets: WNGen and FBGen. Experiments show that the model is SOTA on benchmarks, and flexible enough to evolve without retraining as the knowledge graph grows.","url_abs":"https://arxiv.org/abs/2110.12341v1","url_pdf":"https://arxiv.org/pdf/2110.12341v1.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":"scalable-knowledge-base-completion-with","repo_url":"https://github.com/matthiasrlalisse/hmemnetworks","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"knowledge-base-completion","task_name":"Knowledge Base Completion"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}