Patents by Inventor Tan Yu

Tan Yu has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Publication number: 20260277949
    Abstract: In various examples, methods, systems, processors, and/or machine-readable mediums implement a conversational artificial intelligence (AI) that obtains results from other conversational AIs, and generates output based at least on those results. In at least one embodiment, an orchestrator conversational AI receives input(s), and uses one or more strategies to obtain one or more results from one or more expert conversational AIs based in part on one or more inputs received by the orchestrator conversational AI, and determines at least one response to the one or more inputs based at least on a portion of the one or more results.
    Type: Application
    Filed: March 14, 2025
    Publication date: September 17, 2026
    Inventors: Lu An, Anbang Xu, Rama Akkiraju, Di Xu, Deepak Bora, Tan Yu, Sidney Knowles, Ashok Marannan, Sean Javiya, Meghana Puvvadi, Abhinav Balasubramanian, Yang Jiao, Aaditya Ramshankar Shukla, Lei Yang, Amanda Burgmeier, Divyanshi Yadav, Hridhay Mehta, Vishal Seth, Tamar Bar, Jasmine Jaksic, Vishnu Vardhaanan Leelakrishnan, Paramaguru Muthusamy, Ashwinikumar Binodnath Jha, Murali Maram
  • Publication number: 20260277980
    Abstract: In various examples, an enhanced retrieval-augmented generation (RAG) system may provide relevant responses to user queries by prioritizing contextually relevant information based on author-related metadata. For instance, the systems and methods of the present disclosure may enrich documents—or portions (e.g., “chunks”) thereof—with metadata indicating an author's role, expertise, or association with a customer or topic. Upon receiving a query, the RAG system may retrieve content using keyword and/or semantic search, and prioritize results based on expert relevance. In some examples, threshold-based filtering or expert relationship graphs may be used to improve retrieval quality and reduce processing costs. Retrieved documents may, in some instances, be broken into smaller parts and summarized in parallel, with intermediate summaries recursively condensed to fit within a language model's context window. A final answer may then be generated using the user query and the refined summary.
    Type: Application
    Filed: December 1, 2025
    Publication date: September 17, 2026
    Applicant: NVIDIA Corporation
    Inventors: Abhinav Balasubramanian, Rama Kalyani Akkiraju, Anbang Xu, Tan Yu, Shivakesh Reddy Annepally, Paramaguru Muthusamy, Lei Yang, Pritam Gundecha, Nicholas Burnett, Meghana Puvvadi, Deepak Bora
  • Publication number: 20260277936
    Abstract: Embodiments of the present disclosure relate to AI-based attribution of issues to specific data processing pipeline stages. For instance, some embodiments detect and classify inefficiency issues, anomaly issues, or failure issues in a data processing pipeline, such as those used in conversational AI systems, search engines, or retrieval-augmented generation (RAG) frameworks. The system monitors various stages of the pipeline to detect the issues that impact the quality, relevance, or performance of the output. By leveraging feedback signals, a language model, and/or correlation techniques via the use of application logs and/or traces, some embodiments automatically attribute issues to specific pipeline stages and provide detailed diagnostic insights.
    Type: Application
    Filed: March 12, 2025
    Publication date: September 17, 2026
    Inventors: Abhinav BALASUBRAMANIAN, Rama Kalyani Tirumala AKKIRAJU, Anbang XU, Aaditya Ramshankar SHUKLA, Pritam Surehlal GUNDECHA, Sidney Anne KNOWLES, Lu AN, Di XU, Yang JIAO, Tony Wijaya SALIM, Japinder SINGH, Tan YU, Meghana PUVVADI, Ryan Richard ANGILLY, Santiago POMBO, Lei YANG, Deepak BORA, Meenakshi MADUGULA
  • Patent number: 12400418
    Abstract: An image segmentation method includes: obtaining an image to be segmented containing a target object; performing at least one image feature fusion processing on associated feature points based on the image to be segmented, and extracting global feature information during each image feature fusion processing, wherein the associated feature points are at least two feature points having a location association relation; and determining, based on the global feature information extracted, a segmentation mask for the target object.
    Type: Grant
    Filed: December 30, 2022
    Date of Patent: August 26, 2025
    Assignee: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.
    Inventors: Jingyi Yang, Tan Yu, Mingming Sun, Ping Li
  • Patent number: 12147497
    Abstract: Current pretrained vision-language models for cross-modal retrieval tasks in English depend upon on the availability of many annotated image-caption datasets for pretraining to have English text. However, the texts are not necessarily in English. Although machine translation (MT) tools may be used to translate text to English, the performance largely relies on MT's quality and may suffer from high latency problems in real-world applications. Embodiments herein address these problems by learning cross-lingual cross-modal representations for matching images and their relevant captions in multiple languages. Embodiments seamlessly combine cross-lingual pretraining objectives and cross-modal pretraining objectives in a unified framework to learn image and text in a joint embedding space from available English image-caption data, monolingual corpus, and parallel corpus. Embodiments are shown to achieve state-of-the-art performance in retrieval tasks on multimodal multilingual image caption datasets.
    Type: Grant
    Filed: April 7, 2022
    Date of Patent: November 19, 2024
    Inventors: Hongliang Fei, Tan Yu, Ping Li
  • Patent number: 12033078
    Abstract: Systems and methods are disclosed for capturing multiple sequences of views of a three-dimensional object using a plurality of virtual cameras. The systems and methods generate aligned sequences from the multiple sequences based on an arrangement of the plurality of virtual cameras in relation to the three-dimensional object. Using a convolutional network, the systems and methods classify the three-dimensional object based on the aligned sequences and identify the three-dimensional object using the classification.
    Type: Grant
    Filed: August 4, 2023
    Date of Patent: July 9, 2024
    Assignee: SNAP INC.
    Inventors: Yuncheng Li, Zhou Ren, Ning Xu, Enxu Yan, Tan Yu
  • Publication number: 20230376757
    Abstract: Systems and methods are disclosed for capturing multiple sequences of views of a three-dimensional object using a plurality of virtual cameras. The systems and methods generate aligned sequences from the multiple sequences based on an arrangement of the plurality of virtual cameras in relation to the three-dimensional object. Using a convolutional network, the systems and methods classify the three-dimensional object based on the aligned sequences and identify the three-dimensional object using the classification.
    Type: Application
    Filed: August 4, 2023
    Publication date: November 23, 2023
    Inventors: Yuncheng Li, Zhou Ren, Ning Xu, Enxu Yan, Tan Yu
  • Patent number: 11755910
    Abstract: Systems and methods are disclosed for capturing multiple sequences of views of a three-dimensional object using a plurality of virtual cameras. The systems and methods generate aligned sequences from the multiple sequences based on an arrangement of the plurality of virtual cameras in relation to the three-dimensional object. Using a convolutional network, the systems and methods classify the three-dimensional object based on the aligned sequences and identify the three-dimensional object using the classification.
    Type: Grant
    Filed: August 1, 2022
    Date of Patent: September 12, 2023
    Assignee: SNAP INC.
    Inventors: Yuncheng Li, Zhou Ren, Ning Xu, Enxu Yan, Tan Yu
  • Patent number: 11704893
    Abstract: Aspects of the present disclosure involve a system comprising a storage medium storing a program and method for receiving a video comprising a plurality of video segments; selecting a target action sequence that includes a sequence of action phases; receiving features of each of the video segments; computing, based on the received features, for each of the plurality of video segments, a plurality of action phase confidence scores indicating a likelihood that a given video segment includes a given action phase of the sequence of action phases; identifying a set of consecutive video segments of the plurality of video segments that corresponds to the target action sequence, wherein video segments in the set of consecutive video segments are arranged according to the sequence of action phases; and generating a display of the video that includes the set of consecutive video segments and skips other video segments in the video.
    Type: Grant
    Filed: September 2, 2021
    Date of Patent: July 18, 2023
    Assignee: Snap Inc.
    Inventors: Zhou Ren, Yuncheng Li, Ning Xu, Enxu Yan, Tan Yu
  • Publication number: 20230133218
    Abstract: An image segmentation method includes: obtaining an image to be segmented containing a target object; performing at least one image feature fusion processing on associated feature points based on the image to be segmented, and extracting global feature information during each image feature fusion processing, wherein the associated feature points are at least two feature points having a location association relation; and determining, based on the global feature information extracted, a segmentation mask for the target object.
    Type: Application
    Filed: December 30, 2022
    Publication date: May 4, 2023
    Applicant: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.
    Inventors: Jingyi Yang, Tan Yu, Mingming Sun, Ping Li
  • Publication number: 20230034794
    Abstract: Systems and methods are disclosed for capturing multiple sequences of views of a three-dimensional object using a plurality of virtual cameras. The systems and methods generate aligned sequences from the multiple sequences based on an arrangement of the plurality of virtual cameras in relation to the three-dimensional object. Using a convolutional network, the systems and methods classify the three-dimensional object based on the aligned sequences and identify the three-dimensional object using the classification.
    Type: Application
    Filed: August 1, 2022
    Publication date: February 2, 2023
    Inventors: Yuncheng Li, Zhou Ren, Ning Xu, Enxu Yan, Tan Yu
  • Publication number: 20220383048
    Abstract: Current pretrained vision-language models for cross-modal retrieval tasks in English depend upon on the availability of many annotated image-caption datasets for pretraining to have English text. However, the texts are not necessarily in English. Although machine translation (MT) tools may be used to translate text to English, the performance largely relies on MT's quality and may suffer from high latency problems in real-world applications. Embodiments herein address these problems by learning cross-lingual cross-modal representations for matching images and their relevant captions in multiple languages. Embodiments seamlessly combine cross-lingual pretraining objectives and cross-modal pretraining objectives in a unified framework to learn image and text in a joint embedding space from available English image-caption data, monolingual corpus, and parallel corpus. Embodiments are shown to achieve state-of-the-art performance in retrieval tasks on multimodal multilingual image caption datasets.
    Type: Application
    Filed: April 7, 2022
    Publication date: December 1, 2022
    Applicant: Baidu USA LLC
    Inventors: Hongliang FEI, Tan YU, Ping LI
  • Patent number: 11410439
    Abstract: Systems and methods are disclosed for capturing multiple sequences of views of a three-dimensional object using a plurality of virtual cameras. The systems and methods generate aligned sequences from the multiple sequences based on an arrangement of the plurality of virtual cameras in relation to the three-dimensional object. Using a convolutional network, the systems and methods classify the three-dimensional object based on the aligned sequences and identify the three-dimensional object using the classification.
    Type: Grant
    Filed: May 8, 2020
    Date of Patent: August 9, 2022
    Assignee: Snap Inc.
    Inventors: Yuncheng Li, Zhou Ren, Ning Xu, Enxu Yan, Tan Yu
  • Publication number: 20210407548
    Abstract: Aspects of the present disclosure involve a system comprising a storage medium storing a program and method for receiving a video comprising a plurality of video segments; selecting a target action sequence that includes a sequence of action phases; receiving features of each of the video segments; computing, based on the received features, for each of the plurality of video segments, a plurality of action phase confidence scores indicating a likelihood that a given video segment includes a given action phase of the sequence of action phases; identifying a set of consecutive video segments of the plurality of video segments that corresponds to the target action sequence, wherein video segments in the set of consecutive video segments are arranged according to the sequence of action phases; and generating a display of the video that includes the set of consecutive video segments and skips other video segments in the video.
    Type: Application
    Filed: September 2, 2021
    Publication date: December 30, 2021
    Inventors: Zhou Ren, Yuncheng Li, Ning Xu, Enxu Yan, Tan Yu
  • Patent number: 11158351
    Abstract: Aspects of the present disclosure involve a system comprising a storage medium storing a program and method for receiving a video comprising a plurality of video segments; selecting a target action sequence that includes a sequence of action phases; receiving features of each of the video segments; computing, based on the received features, for each of the plurality of video segments, a plurality of action phase confidence scores indicating a likelihood that a given video segment includes a given action phase of the sequence of action phases; identifying a set of consecutive video segments of the plurality of video segments that corresponds to the target action sequence, wherein video segments in the set of consecutive video segments are arranged according to the sequence of action phases; and generating a display of the video that includes the set of consecutive video segments and skips other video segments in the video.
    Type: Grant
    Filed: December 20, 2018
    Date of Patent: October 26, 2021
    Assignee: Snap Inc.
    Inventors: Zhou Ren, Yuncheng Li, Ning Xu, Enxu Yan, Tan Yu
  • Publication number: 20210209155
    Abstract: Embodiments of the present disclosure disclose a method and apparatus for retrieving a video, a device and a medium, and relate to the field of data processing technology, and particularly to the field of smart retrieval technology. The method may include: determining, according to a query text and a candidate video, a unified space feature of the query text and a unified space feature of the candidate video based on a conversion relationship between a text semantic space and a video semantic space; determining a similarity between the query text and the candidate video according to the unified space feature of the query text and the unified space feature of the candidate video; and selecting a target video from the candidate video according to the similarity, and using the target video as a query result.
    Type: Application
    Filed: September 16, 2020
    Publication date: July 8, 2021
    Inventors: Yi YANG, Yi LI, Shujing WANG, Jie LIU, Tan YU, Xiaodong CHEN, Lin LIU, Yanfeng ZHU, Ping LI
  • Patent number: 10988735
    Abstract: Described herein are tissues containing semiconductor nanomaterials. In some embodiments, the tissues include vascular cells, cardiomyocytes, and/or cardiac fibroblasts. The tissue may be scaffold-free. In some embodiments, the tissue includes an electrically conductive network. The tissue may exhibit synchronized electrical signal propagation within the tissue. In some embodiments, the tissue exhibits increased functional assembly of cardiac cells and/or increased cardiac specific functions compared to a cardiac tissue prepared using a conventional tissue culture method. Methods of preparing and using such tissues are also described herein.
    Type: Grant
    Filed: January 15, 2016
    Date of Patent: April 27, 2021
    Assignees: Clemson University Research Foundation, MUSC Foundation for Research Development, The University of Chicago
    Inventors: Ying Mei, Tan Yu, Dylan Richards, Donald R. Menick, Bozhi Tian
  • Publication number: 20200356760
    Abstract: Systems and methods are disclosed for capturing multiple sequences of views of a three-dimensional object using a plurality of virtual cameras. The systems and methods generate aligned sequences from the multiple sequences based on an arrangement of the plurality of virtual cameras in relation to the three-dimensional object. Using a convolutional network, the systems and methods classify the three-dimensional object based on the aligned sequences and identify the three-dimensional object using the classification.
    Type: Application
    Filed: May 8, 2020
    Publication date: November 12, 2020
    Inventors: Yuncheng Li, Zhou Ren, Ning Xu, Enxu Yan, Tan Yu
  • Publication number: 20170369847
    Abstract: Provided herein are tissues containing semiconductor nanomaterials and methods of preparing and using the same.
    Type: Application
    Filed: January 15, 2016
    Publication date: December 28, 2017
    Inventors: Ying Mei, Tan Yu, Dylan Richards, Donald R. Menick, Bozhi Tian