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

Researchers developed TRIZ-RAGNER, a retrieval-augmented large language model framework that improves patent analysis and systematic innovation by extracting technical contradictions from patent documents. The system achieved 84.2% F1-score accuracy, outperforming existing methods by 7.3 percentage points through better integration of domain-specific knowledge.

• →TRIZ-RAGNER combines large language models with retrieval-augmented generation to improve patent contradiction mining accuracy.
• →The framework addresses semantic ambiguity and hallucination issues in traditional patent analysis approaches.
• →Testing on the PaTRIZ dataset showed 85.6% precision, 82.9% recall, and 84.2% F1-score performance.
• →The system outperformed GPT-based baselines by 7.3 percentage points in F1-score metrics.
• →The approach reformulates patent contradiction mining as a named entity recognition task with structured knowledge integration.