Patents
US 12724752 B2: System and method for generating thoughts with large language models using codewords
- Granted
- 2026-09-01
- Filed
- 2025-11-21 application 19397339
- Type
- Utility · 20 claims
- Elsewhere
- Google Patents ↗ · USPTO patent document ↗ · USPTO Patent Center record ↗
Abstract
This invention presents an optimized approach for training and operating Large Language Models (LLMs) using codewords. By converting traditional token-based LLMs to codeword-based systems, the method achieves significant efficiency gains. The process involves tokenizing training data and assigning codewords to tokens. LLMs are then trained and operated using these compact codewords instead of conventional tokens. During operation, prompts are converted to codewords, processed by the LLM, and the outputs are converted back to text. This approach reduces the overall cost of training and operating LLMs by approximately, offering a more efficient solution for large-scale language processing tasks.
Source: PTBLXML, file of 2026-09-01 (ipgb20260901.xml).
Assignees
| Assignee | Type | Place | Also on furp.org | Source |
|---|---|---|---|---|
| ATOMBEAM TECHNOLOGIES INC. 213 patents | US company or organization | Moraga, CA | ATOMBEAM TECHNOLOGIES INC. State business registration EXACT high confidence: identical full name and state AtomBeam Technologies Inc. Form D issuer EXACT high confidence: identical full name and state | PTBLXML, file of 2026-09-01 |
Inventors
| Inventor | Place | Patents | Person page | Source |
|---|---|---|---|---|
| Brian Galvin | Silverdale, WA | 128 | PTBLXML, file of 2026-09-01 | |
| Alan McCord | Forney, TX | 36 | PTBLXML, file of 2026-09-01 |
CPC categories
| Subclass | Category | First group | Source |
|---|---|---|---|
| G06F | Electric digital data processing | G06F16/212 | PTBLXML, file of 2026-09-01 |
Cooperative Patent Classification, as currently assigned (one row per subclass).
Assignments and security interests
No recorded assignments or security interests on file for this patent.
USPTO bulk data: PatentsView granted patents, disambiguated (PVGPATDIS) for patents, assignees, inventors, places and CPC categories; the USPTO Patent Assignment Dataset (ECORSEXC) for assignments and security interests. Patents filed in the last 20 years. No claims or full text. Links to other records are by name (and state): EXACT means the identical full name and state; FUZZY means a looser name match, with the confidence shown.