Google patents method to train AI image models using unpaired text data
Google describes a method for training text-to-image and image-to-text models using only unpaired text data. The system generates a synthetic image from text, converts it back to text, and adjusts the models based on how closely the output matches the original input.
In the filing’s own words
A computer-implemented method includes obtaining, by a computing system that may include one or more computing devices, a text-to-image model and an image-to-text model. The method also includes, for each of one or more training iterations: accessing, by the computing system, an unpaired text input; processing, by the computing system, the unpaired text input with the text-to-image model to generate a synthetic image; processing, by the computing system, the synthetic image with the image-to-text model to generate predicted text; evaluating, by the computing system, a loss function that compares the predicted text to the unpaired text input to generate a loss value; and modifying, by the computing system, one or more parameter values of one or both of the text-to-image model and the image-to-text model based on the loss function.
What exactly they claim (claim 1)
1. A computer-implemented method to train modality conversion models using unpaired text data, the method comprising: obtaining, by a computing system comprising one or more computing devices, a text-to-image model and an image-to-text model; for each of one or more training iterations: accessing, by the computing system, an unpaired text input; processing, by the computing system, the unpaired text input with the text-to-image model to generate a synthetic image; processing, by the computing system, the synthetic image with the image-to-text model to generate predicted text; evaluating, by the computing system, a loss function that compares the predicted text to the unpaired text input to generate a loss value; and modifying, by the computing system, one or more parameter values of one or both of the text-to-image model and the image-to-text model based on the loss function.
Facts
- Company
- Google (Google LLC)
- Official title
- Leveraging Unpaired Data for Cross-Modal Generative Models via Cycle Consistency
- Publication number
- US20260278464A1
- Application number
- 19524304
- Filed
- Sep 30, 2024
- Published
- Sep 17, 2026
- Inventors
- 6
- Claims
- 13
Classified under
G06N 20/00 Physics & computingMore from Google in AI & software
- Google patents video search that answers questions using AI descriptions Sep 15, 2026
- Google patents photo tool that removes people and their gear Sep 15, 2026
- Google patents AI that writes live meeting notes and summaries Sep 15, 2026
- Google patents AI tool that turns messy emergency notes into clear messages Sep 3, 2026
- Google patents AI that clicks through apps to finish tasks for you Sep 24, 2026
Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.