Facebook, Inc.

Google Inc.

Nec Corporation

E Ink California, LLC

Applied Materials, Inc.

Midea Group Co., Ltd.

Meta Platforms, Inc.

Xin Liu

Mark E Zuckerberg

Michael Steven Vernal

Yofay Kari Lee

Austin Haugen

Spencer G Ahrens

Ray C He

Chaitanya Mishra

Brian N Kuhn

Yihong Gong

Hui Du

Ming Wang

Shengyue Ji

Chen Zhang

Xiaojun Liang

Haibin Huang

Abhijit Basu Mallick

Robert Jan Visser

Rui Cheng

Fei C Wang

Maher Afif Saba

Bradley David Townsend

Parthiv Bharatbhai Patel

Ludovic Godet

Yixiong Yang

Yang Li

Trystan G Upstill

Kurtis Leschkies

Wei V Tang

James Lee Wogulis

Peng Yang

Yifei Wang

Shyamsundar Rajaram

Asad K Awan

Fei Wang

Robert Oliver Burns Zeldin

Rohan Kuruvilla

Wenyi Liu

Jurgen Anne Francois Marie Van Gael

Stewart Yang

Avi Mehta

Zheng Chen

Nimish Rameshbhai Shah

Matteo Slanina

Yining Wu

Yu Ning

Lixing Huang

Hao Shi

Fei Xie

Bingyue Peng

Dihong Gao

Junfeng Pan

Derrick E Bass

Zhen Yao

Piyush Bansal

Embodiments of the present disclosure relate to processes for filling trenches. The process includes depositing a first amorphous silicon layer on a surface of a layer and a second amorphous silicon layer in a portion of a trench formed in the layer, and portions of side walls of the trench are exposed. The first amorphous silicon layer is removed. The process further includes depositing a third amorphous silicon layer on the surface of the layer and a fourth amorphous silicon layer on the second amorphous silicon layer. The third amorphous silicon layer is removed. The deposition/removal cyclic processes may be repeated until the trench is filled with amorphous silicon layers. The amorphous silicon layers form a seamless amorphous silicon gap fill in the trench since the amorphous silicon layers are formed from bottom up.

Method for Si gap fill by PECVD

A computer-implemented method includes receiving, by an application executing on a computing device, a first environment with a first identifier, in response to a first request comprising an authentication code. The method further includes associating, by the application, the first identifier with the authentication code, sending, by the application, a second request comprising the first identifier and the authentication code, and in response to sending the second request, receiving, by the application, a second environment with a second identifier. The second environment is a version snapshot of the first environment.

Systems and methods for accessing multiple resources via one identifier

An electrophoretic display medium includes a front and a rear electrode, at least one of the front and rear electrodes being transparent, and an encapsulated dispersion fluid containing a plurality of pigments positioned between the front and rear electrode. The plurality of pigments includes a first and a second type of organic pigment particle. The first type of organic pigment particle has a first color and a first charge polarity. The second type of organic pigment particle has a second color different from the first color and a second charge polarity the same as the first charge polarity. At least one of the first and second types of organic pigment particle includes a silica coating and a polymeric stabilizer covalently bonded to the silica coating.

Color organic pigments and electrophoretic display media containing the same

A natural language processing (NLP) apparatus includes a housing, a built-in voice input interface; a built-in data communication interface configured to establish data communication with multiple types of appliances; a built-in NLP module, and a built-in control device. A first voice input is received through the built-in voice input interface; if the target appliance is a first appliance of a first appliance type, the first voice input is processed using a first NLP model of the built-in NLP module to obtain a first machine command, and the first machine command is sent via the built-in data communication interface to the first appliance; and if the target appliance is a second appliance of a second appliance type, the first voice input is processed using a second NLP model of the built-in NLP module, and the second machine command is sent via the built-in data communication interface to the second appliance.

System and method for providing portable natural language processing interface across multiple appliances

Machine-trained models are generated based on a model description that defines parameters for training the model and that can inherit parameters from parent model descriptions. When a parent model description changes, the changes made to the parent model description are applied to the model description automatically. When a target model is re-generated, a description of the set of parameters for generating the target model is received. The parent model is then identified from the received description, and a description of the set of parameters for generating the parent model is retrieved. Using the description for the target model and the parent model, a pipeline for generating the target model is generated. Finally, the pipeline is executed to generate the target model.

Automated generation of machine learning models

An online system provides content items including URLs of third party websites to client devices. The client devices prefetch data from the third party website via the URL. The online system controls the prefetch rate for the client devices based on information received from the client devices. The online system may reduce the prefetch rate if an aggregate number of prefetches from the third party website during a time interval exceeds a prefetch quota. The online system may reduce the prefetch rate if the client devices indicate poor performance of the third party. The online system may determine whether to prefetch data from a URL included in a content item based on a likelihood of the user ignoring the content item if here is a delay in loading of the content item in the client device.

Managing prefetching of content from third party websites by client devices based on prediction of user interactions

A method and system of customizing a portable voice-based control user interface for multiple types of appliances are disclosed. The method performed at a user device includes establishing a data communication connection with a voice control apparatus; detecting a first user request to update a NLP module of the voice control apparatus; establishing a connection to a NLP model server; displaying a listing of appliance types and a respective listing of appliance functions for each appliance type, in a graphical user interface of the user device; receiving user selection of a first set of appliance functions for a first appliance type and a second set of appliance functions for a second appliance type; downloading, from the NLP model server, a first NLP model for the first set of appliance functions for the first appliance type, and a second NLP model for the second set of appliance functions for the second appliance type; and integrating the first NLP model and second NLP model into the NLP module of the voice control apparatus.

System and method for customizing portable natural language processing interface for appliances

An online system obtains a composite prediction associated with a content item indicating a likelihood that a viewing user of the online system will perform a type of conversion associated with the content item via one or more paths of events leading to the type of conversion. The online system obtains the composite prediction based on a tree data structure describing the path(s) of events. Upon identifying an opportunity to present content to the viewing user, the online system identifies the tree data structure corresponding to the type of conversion from multiple tree data structures maintained in the online system and obtains a composite prediction associated with the content item by evaluating and performing operations associated with nodes of the tree data structure while traversing the nodes. The online system then determines whether to present the content item to the viewing user based on the composite prediction.

Obtaining a composite prediction indicating a likelihood that a user of an online system will perform a conversion associated with a content item via one or more paths of events

An electrophoretic display medium includes a front and a rear electrode, at least one of the front and rear electrodes being transparent, and an encapsulated dispersion fluid containing a plurality of pigments positioned between the front and rear electrode. The plurality of pigments includes a first and a second type of organic pigment particle. The first type of organic pigment particle has a first color and a first charge polarity. The second type of organic pigment particle has a second color different than the first color and a second charge polarity the same as the first charge polarity. At least one of the first and second types of organic pigment particle includes a silica coating and a polymeric stabilizer covalently bonded to the silica coating.

Systems and methods for accessing multiple resources via one identifier are described, including receiving an input that includes at least an authentication code from an application; associating the authentication code with a resource; receiving another input, which includes at least an identifier associated with the resource and the authentication code, from the application requesting the resource; determining that the application can access another resource based on the authentication code; and based on the determining, providing the another resource to the application.

A method and system of providing a portable voice-based control user interface for multiple types of appliances are disclosed. The method includes activating a built-in voice communication interface of a voice control apparatus; selecting a first target appliance to receive voice-based commands; receiving a first voice input; in accordance with a determination that the first target appliance is a first appliance of a first appliance type, processing the first voice input using a first NLP model corresponding to the first appliance type to obtain a first machine command, and sending the first machine command to the first appliance; and in accordance with a determination that the first target appliance is a second appliance of a second appliance type, processing the first voice input using a second NLP model corresponding to the second appliance type to obtain a second machine command, and sending the second machine command to the second appliance.

The method of treating a film stack includes depositing a barrier film containing a metal into a via formed within a dielectric layer disposed on a substrate and depositing a metal contact on the barrier film within the via, where a void is located within the barrier film or between the barrier film and the metal contact. The method also includes exposing the metal contact and the barrier film to an oxidizing agent at a temperature of less than 400&#xb0; C. and at a pressure of about 20 bar to about 100 bar within a process chamber to produce a metal oxide within the void.

High pressure annealing of metal gate structures

Growing community of inventors

San Jose, CA, United States of America

Xin Liu