TRIZ-RAGNER: A Retrieval-Augmented Large Language Model for TRIZ-Aware Named Entity Recognition in Patent-Based Contradiction Mining

2026-02-27

AI Summary

This study addresses the limitations of existing approaches in identifying TRIZ contradiction parameters from patent texts, which suffer from semantic ambiguity, domain dependency, and insufficient generalization. To overcome these challenges, the work formulates TRIZ parameter recognition as a semantic-level named entity recognition task and introduces a retrieval-augmented framework that integrates a structured TRIZ knowledge base. Specifically, it employs dense retrieval to fetch relevant knowledge entries, followed by a cross-encoder reranker to enhance retrieval precision, and leverages structured prompts for large language models to improve semantic comprehension. This approach effectively mitigates hallucination and substantially improves the accuracy and consistency of parameter extraction, achieving 85.6% precision, 82.9% recall, and an 84.2% F1 score on the PaTRIZ dataset—outperforming the strongest baseline by 7.3 F1 points.

Technology Category

• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Natural Language Processing: Information Extraction
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

TRIZ-based contradiction mining is a fundamental task in patent analysis and systematic innovation, as it enables the identification of improving and worsening technical parameters that drive inventive problem solving. However, existing approaches largely rely on rule-based systems or traditional machine learning models, which struggle with semantic ambiguity, domain dependency, and limited generalization when processing complex patent language. Recently, large language models (LLMs) have shown strong semantic understanding capabilities, yet their direct application to TRIZ parameter extraction remains challenging due to hallucination and insufficient grounding in structured TRIZ knowledge. To address these limitations, this paper proposes TRIZ-RAGNER, a retrieval-augmented large language model framework for TRIZ-aware named entity recognition in patent-based contradiction mining. TRIZ-RAGNER reformulates contradiction mining as a semantic-level NER task and integrates dense retrieval over a TRIZ knowledge base, cross-encoder reranking for context refinement, and structured LLM prompting to extract improving and worsening parameters from patent sentences. By injecting domain-specific TRIZ knowledge into the LLM reasoning process, the proposed framework effectively reduces semantic noise and improves extraction consistency. Experiments on the PaTRIZ dataset demonstrate that TRIZ-RAGNER consistently outperforms traditional sequence labeling models and LLM-based baselines. The proposed framework achieves a precision of 85.6%, a recall of 82.9%, and an F1-score of 84.2% in TRIZ contradiction pair identification. Compared with the strongest baseline using prompt-enhanced GPT, TRIZ-RAGNER yields an absolute F1-score improvement of 7.3 percentage points, confirming the effectiveness of retrieval-augmented TRIZ knowledge grounding for robust and accurate patent-based contradiction mining.

Problem

• TRIZ contradiction mining
• named entity recognition
• patent analysis
• semantic ambiguity
• domain knowledge grounding

Innovation

• Retrieval-Augmented Generation
• TRIZ
• Named Entity Recognition
• Patent Mining
• Large Language Models