{"schema_version":"1.0","canonical_url":"https://patentable.app/patents/US-11526679","patent":{"patent_number":"US-11526679","title":"Efficient transformer language models with disentangled attention and multi-step decoding","assignee":null,"inventors":[],"filing_date":"2020-06-24T00:00:00.000Z","publication_date":"2022-12-13T00:00:00.000Z","cpc_codes":["G06F","G06N","G06F","G06F","G06N","G06N","G06N"],"num_claims":20,"abstract":"Systems and methods are provided for facilitating the building and use of natural language understanding models. The systems and methods identify a plurality of tokens and use them to generate one or more pre-trained natural language models using a transformer. The transformer disentangles the content embedding and positional embedding in the computation of its attention matrix. Systems and methods are also provided to facilitate self-training of the pre-trained natural language model by utilizing multi-step decoding to better reconstruct masked tokens and improve pre-training convergence."},"analysis":{"summary":null,"layman_explanation":null,"technical_analysis":null,"business_analysis":null,"faqs":null,"topics":[],"tech_cluster":null},"seo":{"title":"Efficient transformer language models with disentangled attention and multi-step decoding","description":"Systems and methods are provided for facilitating the building and use of natural language understanding models. The systems and methods identify a plurality of tokens and use them to generate one or ","keywords":[]},"attribution":{"source":"Patentable","source_url":"https://patentable.app","canonical_url":"https://patentable.app/patents/US-11526679","license":"CC-BY-4.0-like","license_terms":"AI-generated analysis on this page (summary, layman_explanation, technical_analysis, business_analysis, faqs) may be reused with attribution and a visible link back to the canonical URL above. Patent abstracts, claims, and bibliographic data are USPTO public domain.","required_link":"https://patentable.app/patents/US-11526679","citation_suggestion":"Patentable. \"Efficient transformer language models with disentangled attention and multi-step decoding\" (US-11526679). https://patentable.app/patents/US-11526679","copyright_holder":"Nomic Interactive Technology LLC"},"links":{"html":"https://patentable.app/patents/US-11526679","json":"https://patentable.app/api/llm-context/US-11526679","site":"https://patentable.app","llms_txt":"https://patentable.app/llms.txt"},"generated_at":"2026-05-31T06:57:45.894Z"}