![](http://storage.googleapis.com/journals-stmjournals-com-wp-media-to-gcp-offload/2022/10/vector.jpg)
Divya K.K,
![](http://storage.googleapis.com/journals-stmjournals-com-wp-media-to-gcp-offload/2022/10/vector.jpg)
Muhammed Dhanish K,
![](http://storage.googleapis.com/journals-stmjournals-com-wp-media-to-gcp-offload/2022/10/vector.jpg)
Muhammed Rashid T,
![](http://storage.googleapis.com/journals-stmjournals-com-wp-media-to-gcp-offload/2022/10/vector.jpg)
Muhammed Rishan O.K,
![](http://storage.googleapis.com/journals-stmjournals-com-wp-media-to-gcp-offload/2022/10/vector.jpg)
Shabin Muneer,
- Student, Department of Computer Science and Engineering, PA College of Engineering, Mangalore, Karnataka, India
- Student, Department of Computer Science and Engineering, PA College of Engineering, Mangalore, Karnataka, India
- Student, Department of Computer Science and Engineering, PA College of Engineering, Mangalore, Karnataka, India
- Student, Department of Computer Science and Engineering, PA College of Engineering, Mangalore, Karnataka, India
- Student, Department of Computer Science and Engineering, PA College of Engineering, Mangalore, Karnataka, India
Abstract
The challenges of solving complex mathematical problems often hinder efficiency in various scientific and engineering domains. This project proposes an innovative solution to these challenges by integrating automated math solvers with large language model (LLM) retrieval-augmented generation (RAG). The proposed system aims to streamline mathematical problem-solving processes, offering a robust and precise tool for real-time recognition, classification, and solution generation. This work provides a novel method of automating the solution of mathematical problems by combining RAG (Retrieval-Augmented Generation) techniques with a Large Language Model (LLM). The suggested technique integrates outside information sources to improve accuracy and relevance while utilizing the advantages of LLMs to comprehend and provide answers for challenging mathematical problems. The system greatly enhances its capacity to handle a variety of mathematical topics, from fundamental arithmetic to advanced calculus, by accessing examples and contextual information from a curated database. By leveraging advanced algorithms and the computational power of LLMs, the system provides accurate and timely solutions to a wide array of mathematical problems. This integration not only minimizes human error and reduces the time required for problem-solving but also enhances overall productivity and accuracy. The automated math solver system with LLM RAG is poised to revolutionize the approach to mathematical problem-solving across various fields, ensuring increased reliability, efficiency, and innovation.
Keywords: LLM, RAG, langchain, NLP, Mathematical problem
[This article belongs to Current Trends in Signal Processing(ctsp)]
Browse Figures
References
- John A. Erickson, Anthony F. Botelho, Steven McAteer, Ashvini Varatharaj, and Neil T. Heffer-nan. 2020. The Automated Grading of Student Open Responses in Mathematics. In Proceedings of the 10th International Conference on Learning Analytics and Knowledge (LAK ’20), March 23–27, 2020, Frankfurt, Germany. ACM, New York, NY, USA, 10 pages. https://doi. org/10. 1145/ 3375462. 3375523.
- Baral, A. F. Botelho, J. A. Erickson, P. Be-nachamardi, and N. T. Heffernan. Improving automated scoring of student open responses in mathematics. International Educational Data Mining Society, 2021.
- Mengxue Zhang,Sami Baral,Neil Heffer- nan,Andrew Lan, ”Automatic Short Math Answer Grading via In-context Meta-learning”. In arXiv preprint arXiv:2205. 15219v3.
- Urrutia and R. Araya, ”Automatically Detect- ing Incoherent Written Math Answers of Fourth- Graders,” Systems, vol. 11, no. 7, p. 353, Jul. 2023.
- Christopher Ormerod, ”Using language models in the implicit automated assessment of mathemat- ical short answer items”. In arXiv preprint arXiv:2308. 11006v1.
- Jia Tracy Shen,Michiharu Yamashita,Ethan Prihar,Neil Heffernan,Xintao Wu,Ben Graff,Dongwon Lee, ”MathBERT: A Pre- trained Language Model for General NLP Tasks in Mathematics Education,” In arXiv preprint arXiv:2106. 07340v5.
- Baral, Karthik Seetharaman, Anthony F. Botelho,Anzhuo Wang,George Heineman1and Neil T. Heffernan1,”Enhancing Auto-scoring of Student Open Responses in the Presence of Mathematical Terms and Expressions”.
- Zong M, Krishnamachari B. Solving math word problems concerning systems of equations with gpt-3. InProceedings of the AAAI Conference on Artificial Intelligence 2023 Jun 26 (Vol. 37, No. 13, pp. 15972-15979).
- White J, Fu Q, Hays S, Sandborn M, Olea C, Gilbert H, Elnashar A, Spencer-Smith J, Schmidt DC. A prompt pattern catalog to enhance prompt engineering with chatgpt. arXiv preprint arXiv:2302.11382. 2023 Feb 21.
![](http://storage.googleapis.com/journals-stmjournals-com-wp-media-to-gcp-offload/2023/07/5a6d427e-ctsp.webp)
Current Trends in Signal Processing
Volume | 14 |
Issue | 02 |
Received | June 18, 2024 |
Accepted | July 17, 2024 |
Published | August 17, 2024 |