Product Design, Manufacturing & Innovation Resources

关于大型语言模型 (LLM) 的最新出版物和专利

大型语言模型(LLM)

提示:除以下选择外,您还可以搜索和过滤我们的 2 个完整数据库:

> 免费出版物搜索工具 < 按作者、主题、关键词、日期或期刊分类。

> 免费专利搜索工具 < 适用于欧洲专利局提供的英文专利。

这是我们最新精选的有关大型语言模型(LLM)的全球英文出版物和专利,涉及众多科学在线期刊,分类并侧重于大型语言模型、LLM、生成式预训练变换器、预训练、变换器架构、梯度下降、GPT、标记化、生成式模型、自我注意机制、屏蔽语言模型和 MLM。

Expert perspectives on the ecosystem of medical AI oversight in the GenAI era

Published on 2026-05-22 by Stephen Gilbert, Tinglong Dai @NATURE npj

Abstract: npj Digital Medicine, Published online: 22 May 2026; doi:10.1038/s41746-026-02785-1In this third article of a three-article series on interviews with Dr Shantanu Nundy, an advisor in the office of the Commissioner of the US FDA on AI governance and digital health policy, we explore how the oversight of generative AI (GenAI) and large language model (LLM)-enabled medical devices extends beyond the remit of any single regulator, requiring coordination among government agencies, health systems, and[...]


Our summary: The article discusses the complexities of overseeing medical AI in the GenAI era. It emphasizes the need for collaboration among various stakeholders, including regulators and health systems. Additionally, it provides practical guidance for innovators on engaging with the FDA.

medical AI, oversight, generative AI, FDA

Publication

Method and system for dynamically refining input questions for generating insights from a database

Patent published on the 2026-05-21 in US under Ref US20260141185 by LTI MINDTREE LTD [IN] (Selvaraj Jeyapaul [in], S Sreetha [in], K Sreelakshmi [in])

Abstract: A method and system for dynamically refining an input question for generating insights from a database is disclosed. A GUI receives an input from a user. A retrieval module retrieves relevant metadata entities by applying a RAG model to the input. A tokenization module determines tokens within the input based on the relevant metadata entities using an LLM. An intent module identifies an intent associated with the input using a LLM and NLP. A classification module classifies the input question in[...]


Our summary: The system dynamically refines user input questions to generate insights from a database. It employs a retrieval module, tokenization, intent identification, and classification to process the input. A boundary condition check ensures completeness, prompting modifications to enhance the input for accurate insights.

dynamic refinement, database insights, LLM, NLP

Patent

Large language models for microservice fault recovery

Patent published on the 2026-05-21 in US under Ref US20260141172 by SAP SE [DE] (Li Hui [cn])

Abstract: In an example embodiment, an LLM is used to identify similar microservices to allow a microservice that is similar to a dependent microservice that is down to be used instead of the downed dependent microservice until the downed dependent microservice can be brought back online. Specifically, the LLM is utilized in two different manners. First, it is used to generate an embedding for an API of each of multiple microservices in a system. These embeddings may then be used to retrieve similar APIs [...]


Our summary: Large language models (LLMs) are employed to identify and select alternative microservices during fault recovery. They generate embeddings for APIs of microservices to retrieve similar APIs when a dependent microservice is down. The LLMs further assess these alternatives based on functionality and parameters to choose a temporary replacement API.

large language models, microservices, fault recovery, API embeddings

Patent

Information processing device, information processing method, program, and recording medium

Patent published on the 2026-05-21 in WO under Ref WO2026105251 by SOFTBANK CORP [JP] (Tamura Shun [jp])

Abstract: In order to provide security measures spanning a plurality of protocols, an information processing device comprises: an acquisition unit that acquires a communication packet including a protocol stack; a tokenization unit that generates a plurality of tokens by tokenizing at least a portion of the communication packet; and an evaluation unit that inputs the plurality of tokens into a model trained through machine learning, and evaluates the security of the communication packet by referring to th[...]


Our summary: The information processing device acquires communication packets and tokenizes them. It generates multiple tokens for evaluation. The evaluation unit uses a machine learning model to assess the security of the packets.

information processing, security measures, tokenization, machine learning

Patent

Llm-assisted audio synthesis framework

Patent published on the 2026-05-21 in US under Ref US20260141914 by BOSCH GMBH ROBERT [DE] (Lin Wei-cheng [us], Wu Ho-hsiang [us], Bondi Luca [us], Ghaffarzadegan Shabnam [us], Kumar Abinaya [us], Das Samarjit [us])

Abstract: Training of an audio foundation model (AFM) is performed using a dataset constructed using low-level audio property control and high-level composition planning. A plurality of digital audio compositions are generated, using a large language model (LLM) as a planner agent. The planner agent is prompted to generate composition plans defining logical combinations of foreground and background digital sounds, event occurrences within the compositions, and digital sound properties. The foreground and [...]


Our summary: An audio synthesis framework utilizes a large language model to generate composition plans for digital audio. The framework trains an audio foundation model with audio-text pairs derived from generated compositions and their descriptions. It ensures high audio quality and logical sound combinations in the digital compositions.

audio synthesis, foundation model, language model, composition planning

Patent

Intelligent dosing platform system with prior authorization and pharmacy benefit manager integration

Patent published on the 2026-05-21 in US under Ref US20260137868 by DATADOSE LLC [US] (Corbett Jeremy [us])

Abstract: [0000] An intelligent dosing platform system for prior authorization and pharmacy benefit management comprising intelligent injection devices with prior authorization modules configured to receive medication release approval from pharmacies, distributing agents, and payor entities. The system analyzes authorization requirements by evaluating individual payor policies, pharmacy benefit manager formulary restrictions, step therapy requirements, medical necessity criteria, seasonality factors, and [...]


Our summary: The system integrates intelligent dosing devices with prior authorization modules for medication approval. It evaluates individual payor policies and formulary restrictions to determine authorization outcomes. An action coordinator manages downstream processes and ensures financial reconciliation between approvals and medication fulfillment.

dosing platform, prior authorization, pharmacy benefit management, automated decision engine

Patent

Artificial intelligence-driven system for validation of livestreamed event content for dispute resolution

Patent published on the 2026-05-21 in US under Ref US20260143177 by RYVAL TECH INC [US] (Tsao Steve Shute [ph], St Johns Mimi [us])

Abstract: [0000] Computer-implemented methods and systems for validating outcomes in livestreamed video events are provided. The methods involve receiving a live video stream, generating a broadcast feed and a parallel analysis feed, selecting video frames at a cadence, and adding timestamps and checksums to the selected frames. An artificial intelligence (AI) vision model detects irregularities in the video frames, and an AI optical character recognition (OCR) or vision-LLM model determines an outcome st[...]


Our summary: The system validates outcomes in livestreamed events using AI-driven methods. It generates a broadcast feed and analyzes video frames for irregularities. An outcome record is created and sent for display, aiding in dispute resolution.

AI validation, livestreamed events, video analysis, dispute resolution

Patent

capability-based monitoring

Published on 2026-05-15 by Katherine C. Kellogg, Bingyang Ye, Yifan Hu, Guergana K. Savova, Byron Wallace, Danielle S. Bitterman @NATURE npj

Abstract: npj Digital Medicine, Published online: 15 May 2026; doi:10.1038/s41746-026-02740-0Large language models (LLMs) have been rapidly adopted in healthcare, but oversight strategies are lacking. We propose capability-based monitoring, motivated by the fact that LLMs are generalist systems whose overlapping internal capabilities are reused across numerous downstream tasks. This approach organizes monitoring around shared capabilities to enable cross-task detection of systemic weaknesses, long-tail er[...]


Our summary: Capability-based monitoring addresses oversight gaps in large language models used in healthcare. It organizes monitoring around shared capabilities to detect systemic weaknesses and errors. The approach provides guidance for developers, organizational leaders, and policymakers.

capability-based monitoring, large language models, healthcare oversight, systemic weaknesses

Publication

涵盖的主题: 大型语言模型、LLM、生成式预训练变换器、预训练、变换器架构、梯度下降、GPT、标记化、生成式模型、自我注意机制、屏蔽语言模型、MLM、ISO/IEC 30170、ISO/IEC 27001、ISO/IEC 25000、ISO/IEC 30164 和 ISO/IEC 27002。

历史背景

1965
1970
1970
1974-11-15
1980
1980
1980
1964
1968
1970
1970
1975
1980
1980
1980

(如果日期未知或不相关,例如“流体力学”,则提供其显著出现的近似估计)

只有注册会员才能免费获得 100% 的全尺寸图片和下载。.

> 登录 <