MODEL TRAINING OF ACTOR MODEL AND CRITIC MODEL

    公开(公告)号:US20250061339A1

    公开(公告)日:2025-02-20

    申请号:US18936628

    申请日:2024-11-04

    Applicant: Lemon Inc.

    Abstract: Embodiments of the present disclosure provide a solution for model training. A method comprises: performing training of a critic model and training of an actor model according to an alternating scheme. The actor model is configured to generate a response for an input question based on a feedback generated by the critic model, and the critic model is configured to generate a feedback to a response generated by the actor mode.

    MACHINE LEARNING MODEL ALIGNMENT
    2.
    发明申请

    公开(公告)号:US20250021891A1

    公开(公告)日:2025-01-16

    申请号:US18900432

    申请日:2024-09-27

    Abstract: A method is proposed for machine learning (ML) model alignment. In the method, a first number of samples is generated by a target ML model based on samples selected from a set of samples. A sample comprises a question-answer pair. The set of samples is updated by adding at least a portion of the first number of samples to the set of samples. The target ML model is trained with at least a portion of the updated set of samples. In this way, the ML model self-generalization ability is unlocked to perform alignment with near-zero human supervision.

    METHOD, APPARATUS, DEVICE FOR CREATING A PLUG-IN AND STORAGE MEDIUM

    公开(公告)号:US20240411526A1

    公开(公告)日:2024-12-12

    申请号:US18812878

    申请日:2024-08-22

    Abstract: According to an embodiment of the disclosure, methods, apparatuses, devices, and storage medium for creating plug-ins are provided. The method includes: providing a plug-in creation portal in a first page associated with an interaction window between a digital assistant and a first user; providing a second page for creating a target plug-in based on a selection of the plug-in creation portal; obtaining inputted plug-in creation information about the target plug-in via the second page; and in response to receiving an operation of releasing the target plug-in, releasing the target plug-in based on the plug-in creation information, the released target plug-in being selectable for an interaction between a user and the digital assistant. In this way, the user can conveniently and quickly complete the creation of the plug-in, thereby helping the user to efficiently complete diversified operations by means of the plug-in during the interaction with the digital assistant.

    Hybrid data processing system and method

    公开(公告)号:US12147432B2

    公开(公告)日:2024-11-19

    申请号:US17462998

    申请日:2021-08-31

    Applicant: LEMON INC.

    Abstract: The present disclosure describes hybrid transactional and analytical processing (HTAP) techniques. A HTAP system comprises a first processing engine configured to perform online transactional processing, a second processing engine configured to perform online analytical processing, and a storage in communication with the first processing engine and the second processing engine. The first processing engine, the second processing engine, and the storage may be modularized and configured to be decoupled from each other. The system may be configured to capture data by the first processing engine in real time, organize the data in a first format in a first part of the storage for use by the first processing engine, propagate the data to a second part of the storage subsystem, and organize the data in a second format in the second part of the storage for use by the second processing engine.

    METHOD, APPARATUS, ELECTRONIC DEVICE AND MEDIUM FOR DETERMINING FAIRNESS IMPACT OF MODEL

    公开(公告)号:US20250013887A1

    公开(公告)日:2025-01-09

    申请号:US18746832

    申请日:2024-06-18

    Applicant: Lemon Inc.

    Abstract: Embodiments of the present disclosure relate to a method, apparatus, electronic device, and medium for determining fairness impact of a sample on a model. The method comprises generating a counterfactual sample by adjusting an original sample in an original sample set, the original sample set being used for generating an original model for performing a classification task. The method further comprises determining a fairness metric of the original model on a validation sample set. In addition, the method further comprises determining fairness impact of the original sample on the original model based on the fairness metric, the original sample, and the counterfactual sample.

    IMAGE PROCESSING
    9.
    发明申请

    公开(公告)号:US20240428552A1

    公开(公告)日:2024-12-26

    申请号:US18821791

    申请日:2024-08-30

    Applicant: Lemon Inc.

    Abstract: There are proposed methods, devices, and computer program products for image processing. In the method, a reference dataset is obtained, the reference dataset comprising a plurality of reference samples, a reference sample in the plurality of reference samples comprising: a reference image and a reference label corresponding to the reference image, the reference label indicating a processing result of the image processing. An influence of the reference dataset on a loss is determined, the loss being used for updating an image model associated with the image processing based on the plurality reference samples. A hyperparameter is determined for updating the image model based on the influence of the reference dataset. The image model is updated based on the hyperparameter, the loss, and the plurality of reference samples. Therefore, the image model may be updated in an accurate and effective way.

    HYBRID DATA PROCESSING SYSTEM AND METHOD

    公开(公告)号:US20230066540A1

    公开(公告)日:2023-03-02

    申请号:US17462998

    申请日:2021-08-31

    Applicant: LEMON INC.

    Abstract: The present disclosure describes hybrid transactional and analytical processing (HTAP) techniques. A HTAP system comprises a first processing engine configured to perform online transactional processing, a second processing engine configured to perform online analytical processing, and a storage in communication with the first processing engine and the second processing engine. The first processing engine, the second processing engine, and the storage may be modularized and configured to be decoupled from each other. The system may be configured to capture data by the first processing engine in real time, organize the data in a first format in a first part of the storage for use by the first processing engine, propagate the data to a second part of the storage subsystem, and organize the data in a second format in the second part of the storage for use by the second processing engine.

Patent Agency Ranking