Daily Arxiv

This is a page that curates AI-related papers published worldwide.
All content here is summarized using Google Gemini and operated on a non-profit basis.
Copyright for each paper belongs to the authors and their institutions; please make sure to credit the source when sharing.

Perovskite-LLM: Knowledge-Enhanced Large Language Models for Perovskite Solar Cell Research

Created by
  • Haebom

Author

Xiang Liu, Penglei Sun, Shuyan Chen, Longhan Zhang, Peijie Dong, Huajie You, Yongqi Zhang, Chang Yan, Xiaowen Chu, Tong-yi Zhang

Outline

To address the need for efficient knowledge management and reasoning systems driven by the rapid advancement of perovskite solar cell (PSC) research, this paper presents a comprehensive knowledge enrichment system that integrates three key components. First, we develop Perovskite-KG, a domain-specific knowledge graph comprising 23,789 entities and 22,272 relationships constructed from 1,517 research papers. Second, we generate two complementary datasets: Perovskite-Chat, consisting of 55,101 high-quality question-answer pairs generated using a novel multi-agent framework, and Perovskite-Reasoning, containing 2,217 carefully curated materials science problems. Third, we introduce two specialized large-scale language models: Perovskite-Chat-LLM for domain-specific knowledge support and Perovskite-Reasoning-LLM for scientific reasoning tasks. Experimental results demonstrate that the proposed system significantly outperforms existing models in both domain-specific knowledge retrieval and scientific reasoning tasks, providing researchers with an effective tool for literature review, experimental design, and complex problem solving in PSC research.

Takeaways, Limitations

•
Takeaways:
◦
We provide knowledge graphs, question-answering datasets, scientific reasoning problem datasets, and large-scale language models specialized for perovskite solar cell research, significantly improving research efficiency.
◦
It demonstrates superior performance over existing models in domain-specific knowledge retrieval and scientific reasoning tasks.
◦
Contributes to the advancement of PSC research by providing effective tools for literature review, experimental design, and complex problem solving.
•
Limitations:
◦
The performance of current systems depends on the quality and quantity of data used, and data bias can affect the results.
◦
As new research findings continue to emerge, continuous updating and management of knowledge graphs and datasets is necessary.
◦
Because it is a model specialized for a specific domain, it may be difficult to apply it to other fields.
◦
Further research on the model's explainability is needed.
👍
Made with Slashpage