IP Library › Granted Patent US 12,462,592
Granted Patent B2
US 12,462,592 · App. 18/160,664 · Granted Nov 4, 2025

Systems and methods for a vision-language pretraining framework

Inventors: Junnan Li (Singapore, SG); Chu Hong Hoi (Singapore, SG)
Assignee: Salesforce, Inc.
G06V20/70G06F40/10G06F40/126G06F40/284G06F40/35G06F40/40G06N20/00G06T9/00G06V10/74G06V10/764G06V10/774
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,462,592
App. No.
18/160,664
Granted
Nov 4, 2025
Kind
B2
Abstract

Embodiments described herein provide a multimodal vision-language model. The multimodal vision-language model contains a Generalist Multimodal Transformer capable of complete multiple tasks using the same set of parameters learning from pre-training. The Generalist Multimodal Transformer allows alignment between frozen, unimodal encoders, such as image encoders and large language models. The Generalist Multimodal Transformer eliminates the need for fine-tuning the image encoders and large language models.

Claims (84)

1 . A method for pre-training a multimodal framework for vision-language tasks, the method comprising:

receiving, from a communication interface, an image and a text accompanying the image;

encoding, by an image encoder, the image into an image representation;

transforming, by a query transformer, the image representation and a set of queries into a transformed representation;

generating, by the query transformer, a text representation based at least in part from the text;

training the query transformer according to one or more vision-language training objectives computed based on the transformed representation and the text representation while keeping the image encoder frozen;

generating, by a pretrained language model, a decoded output text based on an output representation from the updated query transformer;

computing a loss based on the decoded output text and the text accompanying the image; and

training the query transformer based on the loss while keeping the image encoder and the pretrained language model frozen, wherein the pretrained language model includes a text decoder, and wherein the generating, by the pretrained language model, the decoded output text based on the output representation from the updated query transformer comprises:

projecting, via a fully connected layer, the output representation to a same dimension with the pretrained language model; and

generating, by the text decoder, the decoded output text based on the projected output representation.

2 . The method of claim 1 , wherein the set of queries are a set of learnable embeddings, and wherein the transforming, by the query transformer, the image representation and the set of queries into the transformed representation includes:

generating, at one or more attention layers of an image transformer in the query transformer, query embeddings from the set of queries.

3 . The method of claim 2 , wherein the one or more vision-language training objectives comprises an image-text matching objective that is generated by:

applying a self-attention mask between the set of queries and the text to generate the query embeddings;

generating, via a classifier head, a match prediction indicating whether the image and the text are a matching pair based on the query embeddings; and

computing the image-text matching objective based on the match prediction and a ground truth.

4 . The method of claim 2 , wherein the one or more vision-language training objectives comprises an image-text contrastive learning objective that is generated by:

computing an image-text similarity based on the query embeddings and the text representation; and

computing the image-text contrastive learning objective based on the image-text similarity.

5 . The method of claim 2 , wherein the one or more vision-language training objectives comprises an image-grounded text generation objective that is generated by:

applying a multi-modal self-attention mask to the set of queries and the text;

generating a predicted text conditioned on image features based on the applied multi-modal self-attention mask; and

computing the image-grounded text generation objective based on the predicted text and the text.

6 . The method of claim 1 , wherein the training the query transformer according to one or more vision-language training objectives comprises:

updating parameters of the query transformer via backpropagation based on any joint combination of the one or more vision-language training objectives.

7 . The system of claim 1 , wherein the decoded output text is generated token by token by the pretrained language model conditioned on previously generated tokens.

8 . The method of claim 1 , wherein the pretrained language model includes a text encoder and a text decoder, and wherein the generating, by the pretrained language model, the decoded output text based on the output representation from the updated query transformer comprises:

projecting, via a fully connected layer, the output representation to a same dimension with the pretrained language model;

encoding, via the text encoder, the projected output representation prepended to a prefix text into a prefix representation;

decoding, via the text decoder, a suffix text from the prefix representation; and

concatenating the prefix text and the suffix text into the decoded output text.

9 . The method of claim 1 , wherein the decoded output text is generated token by token by the pretrained language model conditioned on previously generated tokens.

10 . The method of claim 1 , wherein the query transformer is first updated according to the one or more vision-language training objectives, and then updated based on the loss.

11 . The system of claim 1 , wherein the pretrained language model includes a text encoder and a text decoder, and wherein the operation of generating, by the pretrained language model, the decoded output text based on the output representation from the updated query transformer comprises:

projecting, via a fully connected layer, the output representation to a same dimension with the pretrained language model;

encoding, via the text encoder, the projected output representation prepended to a prefix text into a prefix representation;

decoding, via the text decoder, a suffix text from the prefix representation; and

concatenating the prefix text and the suffix text into the decoded output text.

12 . A system for pre-training a multimodal framework for vision-language tasks, the system comprising:

a communication interface receiving an image and a text accompanying the image;

a memory storing an image encoder, a query transformer, a pretrained language model, and a plurality of processor-executable instructions; and

one or more processors executing the instructions to perform operations including:

encoding, by the image encoder, the image into an image representation;

transforming, by the query transformer, the image representation and a set of queries into a transformed representation;

generating, by the query transformer, a text representation based at least in part from the text;

training the query transformer according to one or more vision-language training objectives computed based on the transformed representation and the text representation while keeping the image encoder frozen;

generating, by the pretrained language model, a decoded output text based on an output representation from the updated query transformer;

computing a loss based on the decoded output text and the text accompanying the image; and

training the query transformer based on the loss while keeping the image encoder and the pretrained language model frozen, wherein the pretrained language model includes a text decoder, and wherein the generating, by the pretrained language model, the decoded output text based on the output representation from the updated query transformer comprises:

projecting, via a fully connected layer, the output representation to a same dimension with the pretrained language model; and

generating, by the text decoder, the decoded output text based on the projected output representation.

13 . The system of claim 12 , wherein the set of queries are a set of learnable embeddings, and wherein the transforming, by the query transformer, the image representation and the set of queries into the transformed representation includes:

generating, at one or more attention layers of an image transformer in the query transformer, query embeddings from the set of queries.

14 . The system of claim 13 , wherein the one or more vision-language training objectives comprises an image-text contrastive learning objective that is generated by:

computing an image-text similarity based on the query embeddings and the text representation; and

computing the image-text contrastive learning objective based on the image-text similarity.

15 . The system of claim 13 , wherein the one or more vision-language training objectives comprises an image-grounded text generation objective that is generated by:

applying a multi-modal self-attention mask to the set of queries and the text;

generating a predicted text conditioned on image features based on the applied multi-modal self-attention mask; and

computing the image-grounded text generation objective based on the predicted text and the text.

16 . The system of claim 13 , wherein the one or more vision-language training objectives comprises an image-text matching objective that is generated by:

applying a self-attention mask between the set of queries and the text to generate the query embeddings;

generating, via a classifier head, a match prediction indicating whether the image and the text are a matching pair based on the query embeddings; and

computing the image-text matching objective based on the match prediction and a ground truth.

17 . The system of claim 12 , wherein the operation of training the query transformer according to one or more vision-language training objectives comprises:

updating parameters of the query transformer via backpropagation based on any joint combination of the one or more vision-language training objectives.

18 . A non-transitory processor-readable storage medium storing a plurality of processor-executable instructions for pre-training a multimodal framework for vision-language tasks, the instructions executed by one or more processors to perform operations, the method comprising:

receiving, from a communication interface, an image and a text accompanying the image;

encoding, by an image encoder, the image into an image representation;

transforming, by a query transformer, the image representation and a set of queries into a transformed representation;

generating, by the query transformer, a text representation based at least in part from the text;

training the query transformer according to one or more vision-language training objectives computed based on the transformed representation and the text representation while keeping the image encoder frozen;

generating, by a pretrained language model, a decoded output text based on an output representation from the updated query transformer;

computing a loss based on the decoded output text and the text accompanying the image; and

training the query transformer based on the loss while keeping the image encoder and the pretrained language model frozen, wherein the pretrained language model includes a text decoder, and wherein the generating, by the pretrained language model, the decoded output text based on the output representation from the updated query transformer comprises:

projecting, via a fully connected layer, the output representation to a same dimension with the pretrained language model; and

generating, by the text decoder, the decoded output text based on the projected output representation.

19 . The non-transitory processor-readable storage medium of claim 18 , wherein the set of queries are a set of learnable embeddings, and wherein the transforming, by the query transformer, the image representation and the set of queries into the transformed representation includes:

generating, at one or more attention layers of an image transformer in the query transformer, query embeddings from the set of queries.

20 . The non-transitory processor-readable storage medium of claim 19 , wherein the one or more vision-language training objectives comprises an image-text matching objective that is generated by:

applying a self-attention mask between the set of queries and the text to generate the query embeddings;

generating, via a classifier head, a match prediction indicating whether the image and the text are a matching pair based on the query embeddings; and

computing the image-text matching objective based on the match prediction and a ground truth.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 24, 2023
From: LI, JUNNAN; HOI, CHU HONG
To: SALESFORCE, INC.
Reel/Frame 062793/0685 →
Continuity (2)
Provisional Application 63424413 · Nov 10, 2022
Related Publication 20240161520A1 · May 16, 2024
References Cited (12)
US 20230281400A1 · Wang · 2023 [cited by applicant]
US 20240282094A1 · Tsimpoukelli · 2024 [cited by examiner]
Chen et al,“Subject-driven Text-to-Image Generation via Apprenticeship Learning”, arxiv (Cornell University), Apr. 14, 2023,pp. 1-18, XP093186354, DOI: 10.48550/arxiv.2304.00186 Retrieved from the Internet: URL:https://… [cited by applicant]
Jia et al., “Taming Encoder for Zero Fine-tuning Image Customization with Text-to-Image Diffusion Models”, arxiv.org, Cornell University Library, 201 Olin Library Cornell University Ithaca, NY 14853, Apr. 5, 2023, XP091… [cited by applicant]
Li et al.,“BLIP-Diffusion: Pre-trained Subject Representation for Controllable Text-to-Image Generation and Editing”, arxiv.org,May 24, 2023, XP093180794, DOI: 10.48550/arxiv.2305.14720 Retrieved from the Internet: URL:… [cited by applicant]
Ruiz et al.,“DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation”, arxiv (Cornell University), Mar. 15, 2023, XP093179964, DOI: 10.48550/arxiv.2208.12242 Retrieved from the Internet: URL… [cited by applicant]
International Search Report and Written Opinion for PCT/US2024/027830, dated Jul. 17, 2024, 14 pages. [cited by applicant]
Deyao Zhu et al: “MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models”, arxiv.org, Cornell University Library, 201 Olin Library Cornell University Ithaca, NY 14853, Apr. 20, 2023 (Apr.… [cited by applicant]
Jean-Baptiste Alayrac et al: “Flamingo: a Visual Language Model for Few-Shot Learning”, A arxiv.org, Cornell University Library, 201 Olin Library Cornell University Ithaca, NY 14853, Apr. 29, 2022 (Apr. 29, 2022), XP091… [cited by applicant]
Junnan Li et al: “BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models”, arxiv.org, Cornell University Library, 201 Olin Library Cornell University Ithaca, NY 14853, May… [cited by applicant]
International Search Report and Written Opinion mailed Jul. 19, 2024, International Patent Application No. PCT/US2024/027695, 110 pages. [cited by applicant]
Alayrac et al., “Flamingo: a Visual Language Model for Few-Shot Learning”, DeepMind, Apr. 28, 2022., arXiv: 2204.14198v1, pp. 1-66. [cited by applicant]