公开(公告)号：US20240395367A1

公开(公告)日：2024-11-28

申请号：US18694101

申请日：2022-09-20

Applicant: Lemon Inc.

Inventor： Yuwei YANG ,  Siqi OUYANG ,  Hao ZHOU ,  Lei LI ,  Hang LI

IPC: G16C20/50 ,  G06F30/27

Abstract: According to the embodiments of the disclosure, a method, apparatus, device, storage media, and program products for designing ligand molecules is provided. The methods described herein include: editing a first molecular structure with an editing model to determine a second molecular structure, the editing at least comprising deleting a fragment from the first molecular structure or adding a fragment to the first molecular structure; in response to determining that an evaluation of the second molecular structure is better than the first molecular structure, training the editing model based on the editing, the evaluation at least indicating binding capacity between the second molecular structure and a target molecule; and determining a target structure of the ligand molecule for the target molecule with the trained editing model and based on the second molecular structure.

公开(公告)号：US20240378453A1

公开(公告)日：2024-11-14

申请号：US18658582

申请日：2024-05-08

Applicant: Beijing Youzhuju Network Technology Co., Ltd. ,  Lemon Inc.

Inventor： Yegor KLOCHKOV ,  Jean-Francois TON ,  Ruocheng GUO ,  Yang LIU ,  Hang LI

IPC: G06N3/094

Abstract: A system for removing a concept from a trained neural network for executing a classification task, the system comprising: the trained neural network, wherein the trained neural network comprises a hidden layer; and a classifier applied at a layer of the hidden layer, wherein: the classifier defines a representation vector at the layer of the hidden layer, wherein the representation vector classifies instances of the concept and non-instances of the concept at the layer; the classifier defines a concept activation vector, wherein the concept activation vector is a normal vector to the representation vector and the concept activation vector comprises an adversarial penalty objective to reduce the instances of the concept at the layer; and a loss function of the trained neural network is optimised based on a downstream loss of the classification task and the adversarial penalty objective.

公开(公告)号：US20250156657A1

公开(公告)日：2025-05-15

申请号：US18942104

申请日：2024-11-08

Applicant: Lemon Inc. ,  Beijing Youzhuju Network Technology Co., Ltd.

Inventor： Zhichao HUANG ,  Rong YE ,  Yu Ting KO ,  Qianqian DONG ,  Shanbo CHENG ,  Mingxuan WANG ,  Hang LI

IPC: G06F40/58

Abstract: Embodiments of the present disclosure relate to a method and apparatus for speech translation, an electronic device, and a medium. The method includes obtaining an audio in a source language, where the audio includes a specific type of information. The method further includes obtaining prompt content related to a target language. In addition, the method further includes generating, based on the audio and the prompt content, a target-language text corresponding to the audio, where the target-language text includes a punctuation mark corresponding to the specific type of the information.

公开(公告)号：US20250077980A1

公开(公告)日：2025-03-06

申请号：US18952687

申请日：2024-11-19

Applicant: Lemon Inc. ,  Beijing Youzhuju Network Technology Co., Ltd.

Inventor： Xinbo ZHANG ,  Luong Quoc TRUNG ,  Zhanming JIE ,  Peng SUN ,  Xiaoran JIN, JR. ,  Hang LI

IPC: G06N20/00

Abstract: There are provided a method, an apparatus, a device, and a storage medium for model training. In a method, a target model is fine-tuned using a set of training data, each training data including a sample question and corresponding annotation information, the annotation information including policy information for solving the sample question and answer information of the sample question. At least one sample question in the set of training data is provided to the fine-tuned target model to determine a candidate answer to the at least one sample question. The fine-tuned target model is trained based at least on a comparison between the candidate answer and the answer information of the at least one sample question.

公开(公告)号：US20250061507A1

公开(公告)日：2025-02-20

申请号：US18722879

申请日：2023-01-04

Applicant: BEIJING YOUZHUJU NETWORK TECHNOLOGY CO., LTD. ,  LEMON INC.

Inventor： Yuanshun YAO ,  Chong WANG ,  Hang LI

IPC: G06Q30/0601

Abstract: The present disclosure provides a method for obtaining a recommended explanation, a device, and a computer readable medium. The method includes: generating a recommended item by a recommendation model; calculating a similarity between a plurality of explanatory items and the recommended item; obtaining a predetermined number of explanatory items from the plurality of explanatory items, as a recommended explanation of the recommended item, wherein a similarity between the predetermined number of explanatory items and the recommended item is greater than a similarity between other explanatory items and the recommended item; and outputting identification information of the predetermined number of explanatory items.