公开(公告)号：US20210157880A1

公开(公告)日：2021-05-27

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: G06F17/27 ,  G06F17/24

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.

公开(公告)号：US10949452B2

公开(公告)日：2021-03-16

申请号：US15854320

申请日：2017-12-26

Applicant: ADOBE INC.

Inventor： Balaji Vasan Srinivasan ,  Pranav Ravindra Maneriker ,  Natwar Modani ,  Kundan Krishna

IPC: G06F16/30 ,  G06F16/27 ,  G06F16/33 ,  G06F16/338 ,  G06F16/31 ,  G06F40/151 ,  G06F40/205 ,  G06F40/284

Abstract: Embodiments of the present invention provide systems, methods, and computer storage media directed to facilitating corpus-based content generation, in particular, using graph-based multi-sentence compression to generate a final content output. In one embodiment, pre-existing source content is identified and retrieved from a corpus. The source content is then parsed into sentence tokens, mapped and weighted. The sentence tokens are further parsed into word tokens and weighted. The mapped word tokens are then compressed into candidate sentences to be used in a final content. The final content is assembled using ranked candidate sentences, such that the final content is organized to reduce information redundancy and optimize content cohesion.

公开(公告)号：US10733359B2

公开(公告)日：2020-08-04

申请号：US15248675

申请日：2016-08-26

Applicant: ADOBE INC.

Inventor： Balaji Vasan Srinivasan ,  Rishiraj Saha Roy ,  Niyati Chhaya ,  Natwar Modani ,  Harsh Jhamtani

IPC: G06F40/131 ,  G06F16/335 ,  G06F40/169 ,  G06F40/284

Abstract: Systems and methods provide for expanding user-provided content. User-provided input content is received via a user interface. Content that is relevant to the user-provided input content is identified from a repository of previously-generated content. The identified relevant content is divided into content sub-segments. From the content sub-segments, one or more pieces of candidate content are identified based on each content sub-segment's relevance to the received input content. At least one piece of identified candidate content is provided for display. A selection of one or more pieces of identified candidate content is received, such that the selected piece(s) of identified candidate content is appended to the received input content, thereby expanding the user-provided content.

公开(公告)号：US20200242197A1

公开(公告)日：2020-07-30

申请号：US16262655

申请日：2019-01-30

Applicant: Adobe Inc.

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

IPC: G06F17/27 ,  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：US10534854B2

公开(公告)日：2020-01-14

申请号：US16407596

申请日：2019-05-09

Applicant: Adobe Inc.

Inventor： Saumitra Sharma ,  Kundan Krishna ,  Balaji Vasan Srinivasan ,  Aniket Murhekar

IPC: G06F17/22 ,  G06F17/27 ,  G06N20/00

Abstract: A targeted summary of textual content tuned to a target audience vocabulary is generated in a digital medium environment. A word generation model obtains textual content, and generates a targeted summary of the textual content. During the generation of the targeted summary, the words of the targeted summary generated by the word generation model are tuned to the target audience vocabulary using a linguistic preference model. The linguistic preference model is trained, using machine learning on target audience training data corresponding to a corpus of text of the target audience vocabulary, to learn word preferences of the target audience vocabulary between similar words (e.g., synonyms). After each word is generated using the word generation model and the linguistic preference model, feedback regarding the generated word is provided back to the word generation model. The feedback is utilized by the word generation model to generate subsequent words of the summary.

公开(公告)号：US10409898B2

公开(公告)日：2019-09-10

申请号：US15816976

申请日：2017-11-17

Applicant: Adobe Inc.

Inventor： Saumitra Sharma ,  Kundan Krishna ,  Balaji Vasan Srinivasan ,  Aniket Murhekar

IPC: G06F17/22 ,  G06F17/27 ,  G06N20/00

Abstract: A targeted summary of textual content tuned to a target audience vocabulary is generated in a digital medium environment. A word generation model obtains textual content, and generates a targeted summary of the textual content. During the generation of the targeted summary, the words of the targeted summary generated by the word generation model are tuned to the target audience vocabulary using a linguistic preference model. The linguistic preference model is trained, using machine learning on target audience training data corresponding to a corpus of text of the target audience vocabulary, to learn word preferences of the target audience vocabulary between similar words (e.g., synonyms). After each word is generated using the word generation model and the linguistic preference model, feedback regarding the generated word is provided back to the word generation model. The feedback is utilized by the word generation model to generate subsequent words of the summary.

公开(公告)号：US20250005296A1

公开(公告)日：2025-01-02

申请号：US18342954

申请日：2023-06-28

Applicant: Adobe Inc.

Inventor： Koustava Goswami ,  Srikrishna Karanam ,  Joseph Koonthanam Jose ,  Prateksha Udhayanan ,  Balaji Vasan Srinivasan

IPC: G06F40/40 ,  G06V10/77 ,  G06V10/82

Abstract: The present disclosure relates to systems, methods, and non-transitory computer-readable media that implements a vision language machine learning model to generate text representations of an input digital image from localized context tokens. In particular, in some embodiments, the disclosed systems generate image patch feature representations that represent patches from an input image. Further, in some embodiments, the disclosed systems generate localized context tokens from the image patch feature representations and prompt context tokens. Moreover, in some embodiments, by utilizing the localized context tokens, the disclosed systems generate a text representation by utilizing a text encoder of the vision language machine learning model.

公开(公告)号：US20240394942A1

公开(公告)日：2024-11-28

申请号：US18323029

申请日：2023-05-24

Applicant: Adobe Inc.

Inventor： Anant Shankhdhar ,  Samyak Sanjay Mehta ,  Shreya Singh ,  K. V. Vikram ,  Tripti Shukla ,  Srikrishna Karanam ,  Balaji Vasan Srinivasan ,  Vishwa Vinay ,  Niyati Himanshu Chhaya

IPC: G06T11/60 ,  G06F16/58 ,  G06F40/211 ,  G06V30/418

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods for expanding a digital document including a sequence of informational data via supplemental multimodal digital content. In particular, the system expands digital documents with multimodal granular details to dynamically integrate supplemental in-depth information to the digital document. For example, in response to a selection of a specific portion of a digital document, the system generates expanded multimodal content (e.g., text and image content) for the selected portion of the digital document from external text and image sources. Indeed, the system uses existing content from the digital document to select images and combine the selected images with text into image-text pairs that are textually and visually consistent with the digital document. Moreover, the system expands the digital document by inserting the image-text pairs in connection with the selected portion of the digital document.

公开(公告)号：US12147775B2

公开(公告)日：2024-11-19

申请号：US17501602

申请日：2021-10-14

Applicant: Adobe Inc.

Inventor： Niyati Himanshu Chhaya ,  Udit Kalani ,  Roodram Paneri ,  Sreekanth Reddy ,  Niranjan Kumbi ,  Navita Goyal ,  Balaji Vasan Srinivasan ,  Ayush Agarwal

IPC: G06F40/40 ,  G06F40/216 ,  G06F40/279 ,  G06F40/30 ,  G06F40/44 ,  G06F40/56 ,  G06N3/08 ,  G06N3/09

Abstract: A content generator system receives a request to generate content for a target entity, and one or more keywords. The content generator system retrieves, for the target entity, a current stage identifier linking the target entity to a current stage within a multi-stage objective. The content generator system generates an input vector including the current stage identifier, a target stage identifier, a token embedding comprising the one or more keywords, and a position embedding for each of the one or more keywords, the target stage identifier associated with a target stage within the multi-stage objective different from the current stage. The content generator system generates output text content for the target entity by applying a generative transformer network to the input vector. The content generator system transmits the output text content to a computing device associated with the target entity.

公开(公告)号：US11900056B2

公开(公告)日：2024-02-13

申请号：US18112136

申请日：2023-02-21

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

CPC classification number: G06F40/253 ,  G06F40/166 ,  G06F40/44 ,  G06F18/214 ,  G06F18/217 ,  G06F40/169 ,  G06F40/289 ,  G06F40/56 ,  G06N3/045 ,  G06N3/088 ,  G06N5/022 ,  G06V30/418

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.