System und Verfahren zur Verbesserung von Grossen Sprachmodellen mit Next-Diktionsvorhersageerkennung unter Verwendung Neuartiger Token

Details

Veröffentlichungs-Nr.

EP4773035

Anmeldetag

30. Dezember 2025

Veröffentlichung

8. Juli 2026

Rechtsraum

EP

Abstract

A method for enhancing a large-language model (LLM) with next thought generation is provided. The method includes augmenting training data associated with the LLM based on a set of thought tokens arranged in a structure. The LLM is trained on the augmented training data using a computer to generate a trained LLM. A response to a query received at the trained LLM is determined. The trained LLM predicts a series of next steps as a logical progression following the response based on the set of thought tokens to simulate agent-like behavior.

Anmelder

Firma

ABB SCHWEIZ AG

Land

🇨🇭 Schweiz