Adobe, Inc.

International Business Machines Corporation

Oracle International Corporation

Walter Chang

Franck Dernoncourt

Trung Huu Bui

Michael J Welch

Gregg Darryl Wilensky

Lubomira A Dontcheva

Doo Soon Kim

Geoff Baum

Zhe Lin

Aseem Agarwala

Quan Anh Tran

Gierad P Laput

Nadia Ghamrawi

Sachin Soni

Jonathan Worthen Brandt

Anmol Dhawan

Ashish Duggal

Seokhwan Kim

Quan Hung Tran

Emre Demiralp

Jerry R Scoggins, Ii

Scott David Cohen

Gavin Stuart Peter Miller

Nedim Lipka

Varun Manjunatha

Grayson Squier Lang

David Kuspa

Seunghyun Yoon

David Allen Jared

Lidan Wang

Khoi Pham

Kushal Kafle

Daicho Ito

Ramesh Radhakrishna Manuvinakurike

Chris C Chen

Nham Le

Shone Sadler

Nathan Sakunkoo

Hailin Jin

Kevin Gary Smith

Eunyee Koh

William Brandon George

Chen Fang

Sanjeev Tagra

Zhihong Ding

Rajiv Jain

Michael F Kraley

Hung Hai Bui

Arun N Swami

Amir Pouran Ben Veyseh

Carl Dockhorn

Xiang Chen

Larry Melvin Masinter

Simon Hayhurst

Thomas J Malloy

Sasha Spala

Vineet Sharma

Sean Fitzgerald

Timothy Jeewun Ganter

Sangwoo Cho

Pranjal Daga

Charles Van Winkle

Chan Young Park

Suryadeep Kumar Agrawal

Darshan D Prasad

Sean MacAvaney

Elahe Rahimtoroghi

Hans G Schek

Minhoe Hur

Adyasha Maharana

Jianguo Zhang

Qiuyu Chen

Claudia Veronica Roberts

Xuanli He

Kishore Aradhya

Jacqueline Brixey

Nikos Vlassis

Francisco Garcia

Arman Cohan

Nicholas Digiuseppe

Khalil Mrini

Xiaolei Huang

Logan Lebanoff

Embodiments of the present invention provide systems, methods, and non-transitory computer storage media for parsing a given input referring expression into a parse structure and generating a semantic computation graph to identify semantic relationships among and between objects. At a high level, when embodiments of the preset invention receive a referring expression, a parse tree is created and mapped into a hierarchical subject, predicate, object graph structure that labeled noun objects in the referring expression, the attributes of the labeled noun objects, and predicate relationships (e.g., verb actions or spatial propositions) between the labeled objects. Embodiments of the present invention then transform the subject, predicate, object graph structure into a semantic computation graph that may be recursively traversed and interpreted to determine how noun objects, their attributes and modifiers, and interrelationships are provided to downstream image editing, searching, or caption indexing tasks.

Methods and systems for generating a semantic computation graph for understanding and grounding referring expressions

Systems and methods for color prediction are described. Embodiments of the present disclosure receive an image that includes an object including a color, generate a color vector based on the image using a color classification network, where the color vector includes a color value corresponding to each of a set of colors, generate a bias vector by comparing the color vector to teach of a set of center vectors, where each of the set of center vectors corresponds to a color of the set of colors, and generate an unbiased color vector based on the color vector and the bias vector, where the unbiased color vector indicates the color of the object.

Systems and methods for predicting long-tail colors in images

Systems and methods for natural language processing are described. Embodiments are configured to receive a structured representation of a search query, wherein the structured representation comprises a plurality of nodes and at least one edge connecting two of the nodes, receive a modification expression for the search query, wherein the modification expression comprises a natural language expression, generate a modified structured representation based on the structured representation and the modification expression using a neural network configured to combine structured representation features and natural language expression features, and perform a search based on the modified structured representation.

Scene graph modification based on natural language commands

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

System and methods for a text summarization system are described. In one example, a text summarization system receives an input utterance and determines whether the utterance should be included in a summary of the text. The text summarization system includes an embedding network, a convolution network, an encoding component, and a summary component. The embedding network generates a semantic embedding of an utterance. The convolution network generates a plurality of feature vectors based on the semantic embedding. The encoding component identifies a plurality of latent codes respectively corresponding to the plurality of feature vectors. The summary component identifies a prominent code among the latent codes and to select the utterance as a summary utterance based on the prominent code.

Topical vector-quantized variational autoencoders for extractive summarization of video transcripts

Systems and methods for intent discovery and video summarization are described. Embodiments of the present disclosure receive a video and a transcript of the video, encode the video to obtain a sequence of video encodings, encode the transcript to obtain a sequence of text encodings, apply a visual gate to the sequence of text encodings based on the sequence of video encodings to obtain gated text encodings, and generate an intent label for the transcript based on the gated text encodings.

Multimodal intent discovery system

The present disclosure describes a model for large scale color prediction of objects identified in images. Embodiments of the present disclosure include an object detection network, an attention network, and a color classification network. The object detection network generates object features for an object in an image and may include a convolutional neural network (CNN), region proposal network, or a ResNet. The attention network generates an attention vector for the object based on the object features, wherein the attention network takes a query vector based on the object features, and a plurality of key vector and a plurality of value vectors corresponding to a plurality of colors as input. The color classification network generates a color attribute vector based on the attention vector, wherein the color attribute vector indicates a probability of the object including each of the plurality of colors.

Decompositional learning for color attribute prediction

The present disclosure describes systems and methods for dialog processing and information retrieval. Embodiments of the present disclosure provide a dialog system (e.g., a task-oriented dialog system) with adaptive recurrent hopping and dual context encoding to receive and understand a natural language query from a user, manage dialog based on natural language conversation, and generate natural language responses. For example, a memory network can employ a memory recurrent neural net layer and a decision meta network (e.g., a subnet) to determine an adaptive number of memory hops for obtaining readouts from a knowledge base. Further, in some embodiments, a memory network uses a dual context encoder to encode information from original context and canonical context using parallel encoding layers.

Dialog system with adaptive recurrent hopping and dual context encoding

Systems and methods for natural language processing are described. One or more embodiments of the present disclosure receive a document comprising a plurality of words organized into a plurality of sentences, the words comprising an event trigger word and an argument candidate word, generate word representation vectors for the words, generate a plurality of document structures including a semantic structure for the document based on the word representation vectors, a syntax structure representing dependency relationships between the words, and a discourse structure representing discourse information of the document based on the plurality of sentences, generate a relationship representation vector based on the document structures, and predict a relationship between the event trigger word and the argument candidate word based on the relationship representation vector.

Inducing rich interaction structures between words for document-level event argument extraction

Systems and methods are provided for providing a navigation interface to access or otherwise use electronic content items. In one embodiment, an augmentation application identifies at least one entity referenced in a document. The entity can be referenced in at least two portions of the document by at least two different words or phrases. The augmentation application associates the at least one entity with at least one multimedia asset. The augmentation application generates a layout including at least some content of the document referencing the at least one entity and the at least one multimedia asset associated with the at least one entity. The augmentation application renders the layout for display.

Augmenting text with multimedia assets

The present disclosure provides a memory-based neural network for question answering. Embodiments of the disclosure identify meta-evidence nodes in an embedding space, where the meta-evidence nodes represent salient features of a training set. Each element of the training set may include a questions appended to a ground truth answer. The training set may also include questions with wrong answers that are indicated as such. In some examples, a neural Turing machine (NTM) reads a dataset and summarizes the dataset into a few meta-evidence nodes. A subsequent question may be appended to multiple candidate answers to form an input phrase, which may also be embedded in the embedding space. Then, corresponding weights may be identified for each of the meta-evidence nodes. The embedded input phrase and the weighted meta-evidence nodes may be used to identify the most appropriate answer.

Memory-based neural network for question answering

The present disclosure relates to generating digital responses based on digital dialog states generated by a neural network having a dynamic memory network architecture. For example, in one or more embodiments, the disclosed system provides a digital dialog having one or more segments to a dialog state tracking neural network having a dynamic memory network architecture that includes a set of multiple memory slots. In some embodiments, the dialog state tracking neural network further includes update gates and reset gates used in modifying the values stored in the memory slots. For instance, the disclosed system can utilize cross-slot interaction update/reset gates to accurately generate a digital dialog state for each of the segments of digital dialog. Subsequently, the system generates a digital response for each segment of digital dialog based on the digital dialog state.

Utilizing a dynamic memory network for state tracking

This disclosure relates to methods, non-transitory computer readable media, and systems that can classify term sequences within a source text based on textual features analyzed by both an implicit-class-recognition model and an explicit-class-recognition model. For example, by applying machine-learning models for both implicit and explicit class recognition, the disclosed systems can determine a class corresponding to a particular term sequence within a source text and identify the particular term sequence reflecting the class. The dual-model architecture can equip the disclosed systems to apply (i) the implicit-class-recognition model to recognize implicit references to a class in source texts and (ii) the explicit-class-recognition model to recognize explicit references to the same class in source texts.

Classifying terms from source texts using implicit and explicit class-recognition-machine-learning models

Systems and methods for sentence fusion are described. Embodiments receive coreference information for a first sentence and a second sentence, wherein the coreference information identifies entities associated with both a term of the first sentence and a term of the second sentence, apply an entity constraint to an attention head of a sentence fusion network, wherein the entity constraint limits attention weights of the attention head to terms that correspond to a same entity of the coreference information, and predict a fused sentence using the sentence fusion network based on the entity constraint, wherein the fused sentence combines information from the first sentence and the second sentence.

Learning to fuse sentences with transformers for summarization

Growing community of inventors

San Jose, CA, United States of America

Walter Chang