公开(公告)号：US11636264B2

公开(公告)日：2023-04-25

申请号：US17467672

申请日：2021-09-07

Applicant: Adobe Inc.

Inventor： Balaji Vasan Srinivasan ,  Gaurav Verma ,  Bakhtiyar Hussain Syed ,  Anandhavelu Natarajan

IPC: G06F40/253 ,  G06F40/44 ,  G06F40/166 ,  G06N5/022 ,  G06F40/169 ,  G06F40/289 ,  G06V30/418 ,  G06N3/045 ,  G06F18/21 ,  G06N3/088 ,  G06F18/214 ,  G06F40/56 ,  G06F17/00

Abstract: Rewriting text in the writing style of a target author is described. A stylistic rewriting system receives input text and an indication of the target author. The system trains a language model to understand the target author's writing style using a corpus of text associated with the target author. The language model may be transformer-based, and is first trained on a different corpus of text associated with a range of different authors to understand linguistic nuances of a particular language. Copies of the language model are then cascaded into an encoder-decoder framework, which is further trained using a masked language modeling objective and a noisy version of the target author corpus. After training, the encoder-decoder framework of the trained language model automatically rewrites input text in the writing style of the target author and outputs the rewritten text as stylized text.

公开(公告)号：US20220414400A1

公开(公告)日：2022-12-29

申请号：US17902586

申请日：2022-09-02

Applicant: Adobe Inc.

Inventor： Navita Goyal ,  Balaji Vasan Srinivasan ,  Anandha velu Natarajan ,  Abhilasha Sancheti

IPC: G06K9/62 ,  G06F40/40 ,  G06K9/00 ,  G06V30/414 ,  G06F40/205

Abstract: In some embodiments, a style transfer computing system receives, from a computing device, an input text and a request to transfer the input text to a target style combination including a set of target styles. The system applies a style transfer language model associated with the target style combination to the input text to generate a transferred text in the target style combination. The style transfer language model comprises a cascaded language model configured to generate the transferred text. The cascaded language model is trained using a set of discriminator models corresponding to the set of target styles. The system provides, to the computing device, the transferred text.

公开(公告)号：US11373210B2

公开(公告)日：2022-06-28

申请号：US16830886

申请日：2020-03-26

Applicant: Adobe Inc.

Inventor： Niranjan Shivanand Kumbi ,  Ajay Awatramani ,  Balaji Vasan Srinivasan ,  Reddy Sreekanth ,  Niyati Himanshu Chhaya

IPC: G06F40/279 ,  G06Q30/02 ,  G06F40/30

Abstract: Techniques and systems are described for content interest from interaction information. Keywords are extracted from digital content, and relevance values are determined based on the keywords that captures both the statistical and semantic significance of topics in the digital content through use of a network representation. Interest values for an entity are determined based on the relevance values and an interaction dataset, which capture both the statistical and semantic significance of the topics with respect to the entity. The interest values may be utilized to control output of digital content to a client device.

公开(公告)号：US20220138474A1

公开(公告)日：2022-05-05

申请号：US17090013

申请日：2020-11-05

Applicant: Adobe Inc.

Inventor： Gaurav Verma ,  Balaji Vasan Srinivasan ,  Trikay Nalamada ,  Pranav Goel ,  Keerti Harpavat ,  Aman Mishra

IPC: G06K9/00 ,  G11B27/034 ,  G11B27/34

Abstract: A method for personalized playback of a video as performed by a video platform includes parsing a video into segments based on visual and audio content of the video. The platform creates multimodal fragments that represent underlying segments of the video, and then orders the multimodal fragments based on a preference of a target user. The platform thus enables nonlinear playback of the segmented video in accordance with the multimodal fragments.

公开(公告)号：US20220121879A1

公开(公告)日：2022-04-21

申请号：US17073258

申请日：2020-10-16

Applicant: Adobe Inc.

Inventor： Navita Goyal ,  Balaji Vasan Srinivasan ,  Anandha velu Natarajan ,  Abhilasha Sancheti

IPC: G06K9/62 ,  G06K9/00 ,  G06F40/205

Abstract: In some embodiments, a style transfer computing system generates a set of discriminator models corresponding to a set of styles based on a set of training datasets labeled for respective styles. The style transfer computing system further generates a style transfer language model for a target style combination that includes multiple target styles from the set of styles. The style transfer language model includes a cascaded language model and multiple discriminator models selected from the set of discriminator models. The style transfer computing system trains the style transfer language model to minimize a loss function containing a loss term for the cascaded language model and multiple loss terms for the multiple discriminator models. For a source sentence and a given target style combination, the style transfer computing system applies the style transfer language model on the source sentence to generate a target sentence in the given target style combination.

公开(公告)号：US11194958B2

公开(公告)日：2021-12-07

申请号：US16123966

申请日：2018-09-06

Applicant: Adobe Inc.

Inventor： Pranav Ravindra Maneriker ,  Vishwa Vinay ,  Sopan Khosla ,  Niyati Himanshu Chhaya ,  Natwar Modani ,  Cedric Huesler ,  Balaji Vasan Srinivasan ,  Anandha velu Natarajan

IPC: G06F40/10 ,  G06F40/166 ,  G06N20/00 ,  G06F40/30 ,  G06F40/109 ,  G06F40/103 ,  G06F40/253

Abstract: A fact replacement and style consistency tool is described. Rather than rely heavily on human involvement to replace facts and maintain consistent styles across multiple digital documents, the described change management system identifies factual and stylistic inconsistencies between these documents, in part, using natural language processing techniques. Once these inconsistencies are identified, the change management system generates a user interface that includes indications of the inconsistencies and information describing them, e.g., an indication noting not only a type of inconsistency but also presenting a first portion and at least a second portion of the multiple documents that are factually inconsistent. By automatically identifying these factual and stylistic inconsistencies across multiple documents and presenting indications of such cross-document inconsistencies, the described change management system eliminates human errors in connection with maintaining factual and stylistic consistency over a body of documents.

公开(公告)号：US20210374168A1

公开(公告)日：2021-12-02

申请号：US16888082

申请日：2020-05-29

Applicant: Adobe Inc.

Inventor： Balaji Vasan Srinivasan ,  Sujith Sai Venna ,  Kuldeep Kulkarni ,  Durga Prasad Maram ,  Dasireddy Sai Shritishma Reddy

IPC: G06F16/332 ,  G06F16/35 ,  G06N3/08 ,  G06K9/34 ,  G06F17/18

Abstract: Enhanced techniques and circuitry are presented herein for providing responses to questions from among digital documentation sources spanning various documentation formats, versions, and types. One example includes a method comprising receiving an indication of a question directed to subject having a documentation corpus, determining a set of passages of the documentation corpus related to the question, ranking the set of passages according to relevance to the question, forming semantic clusters comprising sentences extracted from ranked ones of the set of passages according to sentence similarity, and providing a response to the question based at least on a selected semantic cluster.

公开(公告)号：US20210312129A1

公开(公告)日：2021-10-07

申请号：US17348257

申请日：2021-06-15

Applicant: Adobe Inc.

Inventor： Balaji Vasan Srinivasan ,  Kushal Chawla ,  Mithlesh Kumar ,  Hrituraj Singh ,  Arijit Pramanik

IPC: G06F40/284 ,  G06N20/00

Abstract: Certain embodiments involve tuning summaries of input text to a target characteristic using a word generation model. For example, a method for generating a tuned summary using a word generation model includes generating a learned subspace representation of input text and a target characteristic token associated with the input text by applying an encoder to the input text and the target characteristic token. The method also includes generating, by a decoder, each word of a tuned summary of the input text from the learned subspace representation and from a feedback about preceding words of the tuned summary. The tuned summary is tuned to target characteristics represented by the target characteristic token.

公开(公告)号：US20210304253A1

公开(公告)日：2021-09-30

申请号：US16830886

申请日：2020-03-26

Applicant: Adobe Inc.

Inventor： Niranjan Shivanand Kumbi ,  Ajay Awatramani ,  Balaji Vasan Srinivasan ,  Reddy Sreekanth ,  Niyati Himanshu Chhaya

IPC: G06Q30/02 ,  G06F40/279 ,  G06F40/30

Abstract: Techniques and systems are described for content interest from interaction information. Keywords are extracted from digital content, and relevance values are determined based on the keywords that captures both the statistical and semantic significance of topics in the digital content through use of a network representation. Interest values for an entity are determined based on the relevance values and an interaction dataset, which capture both the statistical and semantic significance of the topics with respect to the entity. The interest values may be utilized to control output of digital content to a client device.

公开(公告)号：US11062085B2

公开(公告)日：2021-07-13

申请号：US16694364

申请日：2019-11-25

Applicant: Adobe Inc.

Inventor： Gaurav Verma ,  Balaji Vasan Srinivasan ,  Shiv Kumar Saini ,  Niyati Himanshu Chhaya

IPC: G06F17/00 ,  G06F40/253 ,  G06F40/30 ,  G06F40/166 ,  G06F40/284

Abstract: A method for generating stylistic feature prescriptions to align a body of text with one or more target goals includes receiving, at a stylistic feature model, a body of text, where the body of text is selected by a user via a graphical user interface (GUI). The stylistic feature model identifies stylistic features from the body of text and populates a stylistic feature vector with the stylistic features. A trained de-confounded prediction model receives the stylistic feature vector. The trained de-confounded prediction model using the stylistic feature vector generates a prediction value for each of one or more target goals, compares the prediction value for each of the one or more target goals to a target value for each of the one or more target goals and outputs, for display on the GUI, one or more stylistic feature prescriptions to the body of text based on results of the comparing.