New RAG Framework Enhances LLM-Based Patent Matching Accuracy

Researchers have developed a novel framework for patent matching that leverages self-knowledge retrieval augmented generation (RAG) to improve accuracy. This method guides large language models (LLMs) to autonomously extract key technical entities and build hierarchical ontological structures from patent queries. By integrating FAISS retrieval with a generative matching mechanism, the framework enhances the LLM's understanding of patent innovations, leading to more precise retrieval and matching. Experimental results on real-world patent datasets show the proposed method's effectiveness and potential for intellectual property protection. AI

IMPACT This framework could improve the efficiency and accuracy of intellectual property protection by enhancing LLM capabilities in patent analysis.

RANK_REASON The cluster contains a research paper detailing a new framework for patent matching using LLMs and RAG. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.IR (Information Retrieval) →

• alphaXiv
• arXiv
• CatalyzeX
• Connected Papers
• CORE Recommender
• DagsHub
• Faiss
• Gotit.pub
• Hugging Face
• Influence Flower
• Litmaps
• ScienceCast
• scite Smart Citations

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →