Amazon patents method for using alignment models to pick AI responses
Amazon describes a system where a large language model generates multiple possible answers to a user's request. An alignment model then evaluates these options and selects the best one to send back as the final response.
In the filing’s own words
The application is directed to receiving an alignment model and a prompt to evaluate and generate a response using a large language model. The large language model populates tokens for potential responses based on framework parameters and evaluates the potential responses using an alignment model. The alignment model can be utilized or selected for a prompt at decoding-time and can be used to select a potential response to use to create an output.
What exactly they claim (claim 1)
1.- 20. (canceled)
Facts
- Company
- Amazon (Amazon Technologies, Inc.)
- Official title
- ALIGNMENT MODEL AS DECODING-TIME HEURISTIC
- Publication number
- US20260289158A1
- Application number
- 19558161
- Filed
- Mar 5, 2026
- Published
- Sep 24, 2026
- Inventors
- 8
- Claims
- 21
Classified under
G06F 40/40 Physics & computingG06F 40/284 Physics & computingG06N 20/00 Physics & computingMore from Amazon in AI & software
- Amazon patents system that cites sources for AI answers Sep 29, 2026
- Amazon patents tech to swap words in movies using actor's voice Sep 8, 2026
- Amazon patents ML tool to add graphics to live sports from moving cameras Sep 15, 2026
- Amazon patents shopping assistant that remembers your search filters Oct 6, 2026
- Amazon patents system where AI agents split tasks and talk to each other Oct 20, 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.