AI Research Answer
Patent
5 cited papers · April 11, 2026 · Powered by Researchly AI
I can help you explore the topic of patents! However, your query is quite broad. Based on the retrieved evidence, here is a structured overview of the patent landscape, covering innovation, classification, retrieval, and analysis.
🧠
TL;DR
Patents are legal documents that secure intellectual property rights and play a critical role in technological innovation. The growing complexity of patent docu…
Patents are legal documents that secure intellectual property rights and play a critical role in technological innovation.1Srinivas et al. (2024)1The growing complexity of patent documents and the surge in patent applications have created a pressing need for automated solutions in patent analysis.1
1
Towards Automated Patent Workflows: AI-Orchestrated Multi-Agent Framework for Intellectual Property Management and AnalysisSakhinana Sagar Srinivas, Vijay Sri Vaikunth et al.2024arXiv
View Want to research your own topic? Try it free →
- Patent as an Innovation Indicator — Patent data serve as a unique resource for studying technical change and are widely used as economic indicators of R&D activity.
- Patent Classification — The International Patent Classification (IPC) and Cooperative Patent Classification (CPC) systems organize patents into hierarchical technological domains, enabling structured search and analysis.
- Automated Patent Classification — Deep learning and pre-trained language models such as BERT, XLNet, RoBERTa, and ELECTRA have been applied to multi-label patent classification, achieving state-of-the-art results. Roudsari et al. (2021)
- Patent Retrieval — Summarization-based query methods significantly improve prior-art retrieval effectiveness over conventional approaches that use entire patent sections. Kamateri et al. (2025)
1
Patent Statistics as Economic Indicators: A SurveyZvi Griliches1990National Bureau of Economic Research
View Diagram
Patent Document │ ▼ ┌─────────────────────────────────────────────┐ │ PatExpert Multi-Agent Framework │ │ │ │ ┌──────────┐ ┌──────────────────────┐ │ │ │Meta-Agent│────▶│ Expert Agents │ │ │ └──────────┘ │ - Classification │ │ │ │ │ - Summarization │ │ │ │ │ - Claim Generation │ │ │ ▼ │ - Multi-Patent GRAG │ │ │ ┌──────────┐ └──────────────────────┘ │ │ │ Critique │ │ │ │ Agent │ (Error Handling & Feedback) │ │ └──────────┘ │ └─────────────────────────────────────────────┘ │ ▼ Patent Insights / IP Management Output
Want to research your own topic? Try it free →
Table
| Aspect | Detail |
|---|---|
| Dominant CPC Domains (post-2020) | G06Q30/02 (marketing automation), G06Q10/10 (office automation), G06N20/00 (machine learning) |
| Leading Patent Jurisdiction | United States, followed by WIPO and the European Patent Office |
| Classification Models | BERT, XLNet, RoBERTa, ELECTRA fine-tuned for multi-label patent classification |
| Retrieval Enhancement | Extractive and abstractive summarization used as surrogate queries for prior-art retrieval |
The exponential rise in e-business patents after 2020 reflects a shift from technology adoption toward intellectual property strategy in digital entrepreneurship.1Purnomo et al. (2025)1Patent image retrieval benefits from hierarchical multi-positive contrastive learning that leverages the Locarno International Classification (LIC) taxonomy.2Kavimandan et al. (2025)2
1
Mapping the Patent Landscape of E-Business in Entrepreneurial Innovation: Global Trends and Thematic TaxonomyAgung Purnomo, Gregorius Samodra Bhato Ratu et al.20252025 International Conference on Informatics, Multimedia, Cyber and Information System (ICIMCIS)
View 2
Hierarchical Multi-Positive Contrastive Learning for Patent Image RetrievalKshitij Kavimandan, Angelos Nalmpantis et al.2025arXiv
View Patent classification remains expensive and time-consuming, and the text in patent documents is not always written to efficiently convey knowledge, complicating automated approaches. Roudsari et al. (2021)
Want to research your own topic? Try it free →
- Patents are a unique and critical resource for studying technological change and R&D activity.
- The United States dominates global patent ownership, with significant regional concentration of innovation post-2020. Purnomo et al. (2025)
- Automated multi-agent frameworks like PatExpert can streamline patent classification, summarization, and claim generation.
- Summarization-based queries significantly outperform conventional full-section queries in patent retrieval tasks. Kamateri et al. (2025)
- Hierarchical contrastive learning improves patent image retrieval by leveraging classification taxonomy structures.
1
Patent Statistics as Economic Indicators: A SurveyZvi Griliches1990National Bureau of Economic Research
View 2
Mapping the Patent Landscape of E-Business in Entrepreneurial Innovation: Global Trends and Thematic TaxonomyAgung Purnomo, Gregorius Samodra Bhato Ratu et al.20252025 International Conference on Informatics, Multimedia, Cyber and Information System (ICIMCIS)
View 3
Towards Automated Patent Workflows: AI-Orchestrated Multi-Agent Framework for Intellectual Property Management and AnalysisSakhinana Sagar Srinivas, Vijay Sri Vaikunth et al.2024arXiv
View 4
Hierarchical Multi-Positive Contrastive Learning for Patent Image RetrievalKshitij Kavimandan, Angelos Nalmpantis et al.2025arXiv
View - "Automated patent classification using transformer models BERT XLNet" — to explore deep learning approaches for patent text classification
- "Prior art patent retrieval methods information retrieval" — to understand how patent search systems are evaluated and improved
- "Patent landscape analysis e-business digital entrepreneurship India" — to find India-specific patent trends in digital and technology sectors
Research smarter with AI-powered citations
Researchly finds and cites academic papers for any research topic in seconds. Used by students across India.