Adobe, Inc.

Parade Technologies, Ltd.

Xiang Chen

Ming Qu

Viswanathan Swaminathan

Saayan Mitra

Somdeb Sarkhel

Haijun Zhu

Chi-hsiang Chin

Jian Feng Chen

Walter Chang

Ryan A Rossi

Eunyee Koh

Trung Huu Bui

Atanu Ranjan Sinha

Sungchul Kim

Shiv Kumar Saini

Zhengyu Yuan

Ritwik Sinha

Tong Yu

Niranjan Shivanand Kumbi

Kshitiz Garg

Quan Hung Tran

Seunghyun Yoon

Ali Aminian

Omar Rahman

Uttaran Bhattacharya

Victor Soares Bursztyn

Roel Coppoolse

Han Guo

Sreekanth Reddy

Manas Garg

Jean Bernard Hishamunda

Saurabh Tripathy

Alexandru Ionut Hodorogea

Alan Lai

Yifu Zheng

Swapneel Mehta

Goutham Srivatsav Arra

Jianguo Zhang

Wilbur Tsai

A method, apparatus, non-transitory computer readable medium, and system for document retrieval include obtaining a query and a document. A prompt generator generates a prompt for a reasoning model based on the query and the document. The reasoning model generates a reasoning result based on the prompt. In some cases, the reasoning result indicates that the document answers the query. A machine learning model provides the document in response to the query based on the reasoning result.

In-context and semantic-aware ensemble model for document retrieval

In some embodiments, techniques for identifying email events generated by bot activity are provided. For example, a process may involve applying bot detection patterns to identify bot activity among email response events.

Bot activity detection for email tracking

Systems and methods for natural language processing are described. One or more aspects of a method, apparatus, and non-transitory computer readable medium include receiving a text phrase; encoding the text phrase using an encoder to obtain a hidden representation of the text phrase, wherein the encoder is trained during a first training phrase using self-supervised learning based on a first contrastive loss and during a second training phrase using supervised learning based on a second contrastive learning loss; identifying an intent of the text phrase from a predetermined set of intent labels using a classification network, wherein the classification network is jointly trained with the encoder in the second training phase; and generating a response to the text phrase based on the intent.

Intent detection

Techniques for recommending hashtags, including trending hashtags, are disclosed. An example method includes accessing a graph. The graph includes video nodes representing videos, historical hashtag nodes representing historical hashtags, and edges indicating associations among the video nodes and the historical hashtag nodes. A trending hashtag is identified. An edge is added to the graph between a historical hashtag node representing a historical hashtag and a trending hashtag node representing the trending hashtag, based on a semantic similarity between the historical hashtag and the trending hashtag. A new video node representing a new video is added to the video nodes of the graph. A graph neural network (GNN) is applied to the graph, and the GNN predicts a new edge between the trending hashtag node and the new video node. The trending hashtag is recommended for the new video based on prediction of the new edge.

Open-domain trending hashtag recommendations

Methods and systems disclosed herein relate generally to systems and methods for using a machine-learning model to predict user-engagement levels of users in response to presentation of future interactive content. A content provider system accesses a machine-learning model, which was trained using a training dataset including previous user-device actions performed by a plurality of users in response to previous interactive content. The content provider system receives user-activity data of a particular user and applies the machine-learning model to the user-activity data, in which the user-activity data includes user-device actions performed by the particular user in response to interactive content. The machine-learning model generates an output including a categorical value that represents a predicted user-engagement level of the particular user in response to a presentation of the future interactive content.

Provisioning interactive content based on predicted user-engagement levels

Automatic item placement recommendation is described. An item placement configuration system receives an item for which a recommended placement is to be generated and identifies an entity associated with the item. The item placement configuration system then identifies a multi-domain taxonomy that describes relationships between different entities based on items associated with the different entities published among different domains. A representation of the entity associated with the item to be placed is then identified within the multi-domain taxonomy, along with a representation of at least one similar entity. Upon identifying a similar entity, historic item placement metrics for the similar entity are leveraged to generate a placement recommendation for the received item. In some implementations, the placement recommendation is output with a visual indication of a similar entity and associated performance metrics that were considered in generating the recommended placement.

Automatic item placement recommendations based on entity similarity

This application is directed to transferring data over a data link coupled between two electronic devices. The data link includes a retimer having a full data path and a bit level data path that are coupled in parallel. The data link is initiated with the full data path and a first sequence of data packets is transferred via the full data path in accordance with a low data rate setting. While transferring the first sequence of data packets, the first sequence of data packets is manipulated in the full data path to establish a connection of the data link, and in response to establishing the connection of the data link, the data link is switched from the full data path to the bit level data path.

High speed data links with low-latency retimer

This application is directed to transferring data over a data link coupled between two electronic devices. The data link includes a retimer having a full data path and a bit level data path that are coupled in parallel. A first sequence of data packets is transferred via the bit level data path of the retimer. While transferring the first sequence of data packets, the data link detects initiation of an equalization procedure based on an initiation data packet in the first sequence of data packets. In accordance with detection of the initiation of the equalization procedure, the data link selects the full data path of the retimer for data transfer over the data link. During the equalization procedure, a second sequence of data packets immediately follows the first sequence of data packets, and is transferred via the full data path of the retimer.

Seamless bit-level low-latency retimer for data links

A touch panel display subsystem for bi-directional intra-panel communication includes a forward link and a backward link between a timing controller and one or more source drivers. The forward link includes a timing controller transmitter in the timing controller to receive video signal, video control signal and touch control signal, and to transmit over the forward link in the format of frame based packet, and a source driver receiver in the source driver to receive video signal, video control signal and touch control signal over the forward link. The backward link includes a source driver transmitter in the source driver to receive the touch sensor data and transmit the touch sensor data over the backward link when the backward link is enabled, and a timing controller receiver in the timing controller to receive the touch sensor data over the backward link when the backward link is enabled.

Bi-directional scalable intra-panel interface

Growing community of inventors

San Jose, CA, United States of America

Xiang Chen