Education, Science, Technology, Innovation and Life
Open Access
Sign In

Research on Multimodal Reasoning and Self-Verifying Agents Based on the Brightness Large Model for Report Materials

Download as PDF

DOI: 10.23977/acss.2025.090410 | Downloads: 0 | Views: 54

Author(s)

Jing Xie 1, Shilong Li 1, Chuan Huang 1, Xiangjun Kong 1, Yongjie Zhu 1

Affiliation(s)

1 State Grid Shanghai Electric Power Company Shinan District Power Supply Company, Shanghai, China

Corresponding Author

Jing Xie

ABSTRACT

Manual review of complex State Grid documentation suffers from inefficiency and oversight limitations. To address these issues, this research proposes an intelligent agent framework based on the Brightness Large Model for automated verification. The methodology integrates three core components. First, a shared semantic space fuses text, table, and diagram data to enable deep understanding and structured summarization. Second, a Retrieval-Augmented Generation system maintains a dynamic knowledge base to ensure strict alignment with evolving regulations. Third, a multi-agent pipeline facilitates collaborative rule matching, inconsistency detection, and automated revision. This system provides robust risk warnings and decision support, optimizing resource allocation while advancing smart grid development and national energy security.

KEYWORDS

Multimodal Reasoning, Intelligent Agent, Smart Grid Management

CITE THIS PAPER

Jing Xie, Shilong Li, Chuan Huang, Xiangjun Kong, Yongjie Zhu, Research on Multimodal Reasoning and Self-Verifying Agents Based on the Brightness Large Model for Report Materials. Advances in Computer, Signals and Systems (2025) Vol. 9: 83-90. DOI: http://dx.doi.org/10.23977/acss.2025.090410.

REFERENCES

[1] K. Qian, W. Li, T. Sun, W. Wang, W. Luo, Docrefine: An intelligent framework for scientific document understanding and content optimization based on multimodal large model agents, arXiv preprint arXiv:2508.07021(2025).
[2] Y. Peng, G. Zhang, M. Zhang, Z. You, J. Liu, Q. Zhu, K. Yang, X. Xu, X. Geng, X. Yang, Lmm-r1: Empowering 3b lmms with strong reasoning abilities through two-stage rule-based rl, arXiv preprint arXiv:2503.07536(2025).
[3] M. Dadopoulos, A. Ladas, S. Moschidis, I. Negkakis, Metadata-driven retrieval-augmented generation for financial question answering, arXiv preprint arXiv:2510.24402(2025).
[4] P. Lewis, E. Perez, A. Piktus, F. Petroni, V. Karpukhin, N. Goyal, H. K¨uttler, M. Lewis, W.-t. Yih, T. Rockt¨aschel, et al., Retrieval-augmented generation for knowledge-intensive nlp tasks, Advances in neural information processing systems, 33(2020) 9459–9474.
[5] T. Taipalus, Vector database management systems: Fundamental concepts, use-cases, and current challenges, Cognitive Systems Research, 85 (2024) 101216.
[6] A. Caione, A. L. Guido, A. Martella, R. Paiano, A. Pandurino, Knowledge base support for dynamic information system management, Information Systems and e-Business Management, 14 (3) (2016) 533–576.
[7] H. Ye, Y. Zheng, Y. Li, K. Zhang, Y. Kong, Y. Yuan, Rh-brainfs: regional heterogeneous multimodal brain networks fusion strategy, Advances in Neural Information Processing Systems, 36 (2023) 59286–59303.
[8] J. Tang, T. Fan, C. Huang, Autoagent: A fully-automated and zero-code framework for llm agents, arXiv preprint arXiv:2502.05957(2025).
[9] H. Wang, C. M. Poskitt, J. Sun, Agentspec: Customizable runtime enforcement for safe and reliable llm agents, arXiv preprint arXiv:2503.18666(2025).
[10] J. Juziuk, D. Weyns, T. Holvoet, Design patterns for multi-agent systems: A systematic literature review, in: Agent-Oriented Software Engineering, Springer, 2014, pp. 79–99.
[11] A. Gonz´alez-Briones, F. De La Prieta, M. S. Mohamad, S. Omatu, J. M. Corchado, Multi-agent systems applications in energy optimization problems: A state-of-the-art review, Energies, 11 (8) (2018) 1928.

Downloads: 42205
Visits: 866238

Sponsors, Associates, and Links


All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.