Beijing Youzhuju Network Technology Patente

There are provided methods, devices, and computer program products for image generation, particularly to debiasing text-to-image diffusion models. In a method, a plurality of images is obtained by an image generating model based on a prompt. The plurality of images comprises a plurality of instances of an object, respectively and the object is specified by the prompt. A plurality of attributes of the plurality of instances of the object are determined respectively. The image generating model is updated based on the plurality of attributes and a predetermined distribution of a plurality of predetermined attributes related to the object. With the above method, the images generated by the updated image generating model may follow the predetermined distribution, and the updated image generating model may output debiased results.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for synthetic data generation for service recommendation models. One of the methods includes generating a plurality of synthetic data items, each synthetic data item including user data, context data, service data of a service, and user action data characterizing one or more user actions with respect to the service; for each synthetic data item of the plurality of synthetic data items, processing the synthetic data item using an evaluator to generate a score indicating quality of the synthetic data item; selecting a subset of synthetic data items from the plurality of synthetic data items; and providing the subset of synthetic data items into a service recommendation model for recommending one or more services to one or more users based on the subset of synthetic data items.

An articulated assembly (100), a mechanical joint and a robot. The plurality of structural members (110) of the articulated assembly (100) include a first structural member (111), a second structural member (112), and a third structural member (113). The plurality of bearings (120) of the articulated assembly (100) includes a first bearing (121) and a second bearing (122). Along the axial direction (X), the bearing housing (ST2) of the second structural member (112) is located between the bearing housing (ST1) of the first structural member (111) and the third structural member (113), the first receiving ring (130) is located between the first bearing (121) and the second bearing (122), two end surfaces of the first receiving ring (130) are respectively in abutment with the inner ring (120b) of the first bearing (121) and the inner ring (120b) of the second bearing (122), the second receiving ring (140) is located between the second bearing (122) and the third structural member (113), and two end surfaces of the second receiving ring (140) are respectively in abutment with the inner ring (120b) of the second bearing (122) and the third structural member (113). The articulated assembly (100) further includes a fastening assembly (150) that fastens together the inner ring (120b) of the first bearing (121), the first receiving ring (130), the inner ring (120b) of the second bearing (122), the second receiving ring (140) and the third structural member (113) in the axial direction (X). Thus, the third structural member (113), the inner ring (120b) of the first bearing (121), and the inner ring (120b) of the second bearing (122) can realize linkage.

Embodiments of the disclosure provide a supporting apparatus for a robot and a robot. The supporting apparatus comprises: at least one supporting component, each comprising: a pair of joint assemblies respectively arranged at both ends of the supporting component in an extending direction and respectively rotatably coupled to the first component and the second component of the robot; a pair of driving component arranged between the pair of joint assemblies and respectively comprising a fixing part and a linear driving rod, end parts of the pair of linear driving rods being rotatably coupled to the pair of joint assemblies about the rotation axis; and a pair of force-bearing components arranged to extend in the extending direction and at least partially clamp the pair of joint assemblies, the force-bearing components comprising a pair of hinge points respectively rotatably coupled to the pair of joint assemblies.

Zusammenfassung wird geladen …

Embodiments of the disclosure provide a joint apparatus for a robot and a robot. The joint apparatus includes a base; an output component arranged to be rotatable about mutually perpendicular three axes for three-degree-of-freedom motion, the output component comprising an outer peripheral portion; a first motor arranged on the base and comprising a first output shaft that rotates about a first axis of the three axes; a first transmission assembly having a first input end fixedly coupled to the first output shaft, a first output end rotatably coupled to the outer peripheral portion; and a composite drive assembly fixedly coupled to the first transmission assembly and comprising a second output end and a third output end rotatably coupled to the outer peripheral portion to drive the output component to rotate about a second axis and a third axis of the three axes.

The present application provides a motion control method for legged robots, an electronic device, and a non-transitory computer-readable storage medium. The motion control method includes: obtaining state information, encoded information, and historical actions of the legged robot at a previous moment, as well as first perception information at a current moment; inputting the state information, the encoded information, and the historical actions at the previous moment into a world model to obtain state information at the current moment output from the world model; inputting the state information and the first perception information at the current moment into a motion strategy model to obtain a target action at the current moment output from the motion strategy model; and controlling a motion of the legged robot according to the target action at the current moment.

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

In a method for feature learning, first and second datasets are obtained. The first dataset includes first behavior data, indicating whether an object performs click and conversion behaviors on recommended content, of a first group of objects on a target application. The second dataset includes second behavior data, indicating whether the object performs a click behavior on the recommended content, of a second group of objects on the target application. Based on a union of the first and second datasets, training samples respectively corresponding to objects are generated. Each traning sample includes attribute information of an object and labels. The first and second label respectively indicate whether the object performs a click behavior and whether the object performs a conversion behavior or indicating a predetermined indicator. An object feature extraction model is trained with training samples, and is configured to extract a feature embedding corresponding to the object.

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

According to embodiments of the present disclosure, there is provided a method, an apparatus, a device and a storage medium for interaction. The method comprises: presenting an interactive interface, wherein the interactive interface comprises at least one first interactive object; in response to detecting a selection of a first interactive object in the at least one first interactive object, presenting the selected first interactive object, the prompt information and model parameter associated with the selected first interactive object in the interactive interface; in response to detecting a sending instruction for the selected first interactive object, sending the multimedia content, the prompt information and the model parameter associated with the selected first interactive object to the target machine learning model; and in response to receiving an execution result for the target task from the target machine learning model, presenting the execution result in the interactive interface. Thereby, a user can intuitively know the capability of the machine learning model at a relatively low operation cost.

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …