Entropy-based detection of the fluency of machine-generated text
The entropy-based technique addresses the challenge of selecting fluent LLMs by computing entropy scores, ensuring natural language generation with low entropy models for improved text quality in applications like email responses and code documentation.
WO2026054880A1 Publication Date: 2026-03-12MICROSOFT TECHNOLOGY LICENSING LLC
0 Cites 0 Cited by
Patent Information
- Application Number
- PCT/US2025/037443
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2024-09-05
- Filing Date
- 2025-07-12
- Publication Date
- 2026-03-12
AI Technical Summary
Technical Problem
Machine-generated text often lacks fluency, appearing robotic and unnatural due to high entropy, making it difficult to select the most suitable large language model (LLM) for generating natural language text.
Method used
An entropy-based technique is employed to compute an entropy score for output texts from multiple LLMs, selecting the LLM with low entropy for generating fluent natural language by training an entropy model on fluent training samples.
Benefits of technology
The entropy-based technique effectively identifies and selects the most fluent LLM output, ensuring natural and grammatically correct text generation for applications such as email responses and code documentation.
✦ Generated by Eureka AI based on patent content.
Abstract
An entropy-based technique is used to select a large language model capable of generating fluent natural language text. An entropy model, trained on fluent natural language samples, is used to determine the entropy of a large language model based on an output text generated by the large language model. The entropy of a machine-generated natural language text is used to quantify the amount of information that the large language model holds with respect to the tokens and context of an input text segment. The entropy score of a model is then used to select a large language model capable of generating fluent text or to select the most fluent machine-generated output text produced by a set of large language models.
Need to check novelty before this filing date? Find Prior Art