{"schema_version":"1.0","canonical_url":"https://patentable.app/patents/US-11526774","patent":{"patent_number":"US-11526774","title":"Method for automatically compressing multitask-oriented pre-trained language model and platform thereof","assignee":null,"inventors":[],"filing_date":"2021-12-28T00:00:00.000Z","publication_date":"2022-12-13T00:00:00.000Z","cpc_codes":["G06N","G06N","G06F","G06F","G06N","G06N","G06N","G06N"],"num_claims":10,"abstract":"Disclosed is a method for automatically compressing multi-task oriented pre-trained language model and a platform thereof. According to the method, a meta-network of a structure generator is designed, a knowledge distillation coding vector is constructed based on a knowledge distillation method of Transformer layer sampling, and a distillation structure model corresponding to a currently input coding vector is generated by using the structure generator; at the same time, a Bernoulli distribution sampling method is provided for training the structure generator; in each iteration, each encoder unit is transferred by Bernoulli distribution sampling to form a corresponding coding vector; by changing the coding vector input to the structure generator and a small batch of training data, the structure generator and the corresponding distillation structure are jointly trained, and a structure generator capable of generating weights for different distillation structures can be acquired."},"analysis":{"summary":null,"layman_explanation":null,"technical_analysis":null,"business_analysis":null,"faqs":null,"topics":[],"tech_cluster":null},"seo":{"title":"Method for automatically compressing multitask-oriented pre-trained language model and platform thereof","description":"Disclosed is a method for automatically compressing multi-task oriented pre-trained language model and a platform thereof. According to the method, a meta-network of a structure generator is designed,","keywords":[]},"attribution":{"source":"Patentable","source_url":"https://patentable.app","canonical_url":"https://patentable.app/patents/US-11526774","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-11526774","citation_suggestion":"Patentable. \"Method for automatically compressing multitask-oriented pre-trained language model and platform thereof\" (US-11526774). https://patentable.app/patents/US-11526774","copyright_holder":"Nomic Interactive Technology LLC"},"links":{"html":"https://patentable.app/patents/US-11526774","json":"https://patentable.app/api/llm-context/US-11526774","site":"https://patentable.app","llms_txt":"https://patentable.app/llms.txt"},"generated_at":"2026-05-30T18:50:10.815Z"}