Rule-Guided Transformer Text Generation for Patent Draft Quality
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for generating patent texts lack integration of advanced computer linguistics and Natural Language Processing, leading to inefficiencies in drafting and a need for improved systems and methods to enhance the quality and efficiency of patent document creation.
Innovation Solution
A system combining transformer-based generation with rule-based approaches, including generative adversarial networks, to manage word suggestions, user inputs, and templates, while utilizing user interfaces for collaborative authoring and quality control, enabling the generation of augmented patents with improved formal quality and scope of protection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If transformer-based generation is used for patent text generation, then productivity is improved, but manufacturing precision deteriorates
Solution Approach 1:
The patent merges transformer-based generation with rule-based generation systems. The transformer model handles high-speed text generation while rule-based components ensure quality control through structured templates and constraints specific to patent documentation, thereby achieving both productivity improvement and precision maintenance.
Solution Approach 2:
The system implements feedback mechanisms where generated patent text is evaluated against predefined rules and constraints. The rule-based component provides feedback to the transformer model to correct deviations from patent formatting standards and technical accuracy requirements, ensuring maintained text quality despite automated generation.
2Manufacturing precision
If rule-based generation is used for patent text generation, then manufacturing precision is improved, but productivity deteriorates
Solution Approach 1:
The patent segments the text generation process into multiple stages: initial draft generation by transformer model, intermediate quality checking by rule-based systems, and final refinement. This segmentation allows the rule-based component to focus only on critical quality aspects rather than generating entire documents, thus maintaining precision while improving overall productivity.
3Manufacturing precision
If advanced computer linguistics and NLP are integrated into patent text generation, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The patent employs a unified system architecture where the transformer model serves multiple functions: generating patent text, understanding patent semantics, and evaluating generated content quality. The rule-based component similarly handles multiple tasks including template validation, constraint checking, and formatting verification. This multi-functionality reduces the need for separate specialized components, thereby managing complexity while improving text quality.
Data Source
AI summary
There is disclosed methods and systems for generating text comprising: receiving text input; determining generated text by using a transformer-based generation as an input and/or as an output of one or more rules applied on said transformer-based generation. Described developments relate to aspects comprising implemented logics for rules, composition of rules, transformers and/or adversarial generation networks, types of transformers (finite-memory, infinite-memory), management of words' suggestions and user inputs, management or prior documents, generation by attraction or by repulsion, simplification, generalization, specification, summarization, paraphrasing, heatmaps for essential features, predictions and management of advantages and/or technical effects, management of templates, and collaborative authoring. Various user interfaces are also described.