{"type":"ScholarlyArticle","title":"NeuroSymbolic LLM for Mathematical Reasoning and Software Engineering","authors":[{"name":"Prithwish Jana","slug":"prithwish-jana","given":"Prithwish","family":"Jana","affiliation":"School of Computer Science , Georgia Institute of Technology , Atlanta , USA"}],"abstract":"In recent years, there has been a significant interest in Large Language Models (LLMs) owing to their notable performance in natural language processing (NLP) tasks. However, while their results show promise in mathematical reasoning and software engineering tasks, LLMs have not yet achieved a satisfactory performance level in these domains. In response, current approaches have prioritized scaling up the size of LLMs, necessitating substantial computational resources and data. Our objective, however, is to pursue a different path by developing neurosymbolic language models. We propose to integrate logical and symbolic feedback during the training process, enabling significantly smaller language models to achieve far better reasoning capabilities than the LLMs currently in use.","doi":"10.24963/ijcai.2024/961","url":"https://ijcai.org/proceedings/2024/961","pdf":"https://ijcai.org/proceedings/2024/0961.pdf","markdown":"https://ijcai.org/proceedings/2024/961.md","bibtex":"https://ijcai.org/proceedings/2024/bibtex/961","volume":{"slug":"2024","title":"Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2024","url":"https://ijcai.org/proceedings/2024/","year":2024,"location":"Jeju, Korea","dates":"3-9 August 2024","editors":"Kate Larson","isbn":"978-1-956792-04-1","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2024"},"section":"Doctoral Consortium","track":"Doctoral Consortium","number":961,"pages":{"first":8492,"last":8493},"keywords":["DC: Natural Language Processing","DC: Knowledge Representation and Reasoning","DC: Machine Learning"],"videos":[],"references":[{"position":1,"raw":"[Ganesh et al., 2022] Vijay Ganesh, Sanjit A Seshia, and Somesh Jha. Machine Learning and Logic: A New Fron- tier in Artificial Intelligence. Formal Methods in System Design, 60(3):426-451, 2022.","title":"Machine Learning and Logic: A New Frontier in Artificial Intelligence","authors":"Vijay Ganesh, Sanjit A Seshia, Somesh Jha","year":"2022","venue":"Formal Methods in System Design","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Jana et al., 2024] Prithwish Jana, Piyush Jha, Haoyang Ju, Gautham Kishore, Aryan Mahajan, and Vijay Ganesh. Co- Tran: An LLM-based Code Translator using Reinforce- ment Learning with Feedback from Compiler and Sym- bolic Execution. arXiv:2306.06755, 2024.","title":"Co-Tran: An LLM-based Code Translator using Reinforcement Learning with Feedback from Compiler and Symbolic Execution","authors":"Jana","year":"2024","venue":null,"doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Moura and Ullrich, 2021] Leonardo de Moura and Sebas- tian Ullrich. The Lean 4 Theorem Prover and Program- ming Language. In Automated Deduction-CADE 28, pages 625-635. Springer, 2021.","title":"Leonardo de Moura and Sebastian Ullrich. The Lean 4 Theorem Prover and Programming Language","authors":"Ullrich Moura","year":"2021","venue":"Automated Deduction-CADE 28","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Trinh et al., 2024] Trieu Trinh, Yuhuai Wu, Quoc Le, He He, and Thang Luong. Solving Olympiad Geometry without Human Demonstrations. Nature, 625, 2024.","title":"Solving Olympiad Geometry without Human Demonstrations","authors":"Trieu Trinh, Yuhuai Wu, Quoc Le, He He, Thang Luong","year":"2024","venue":"Nature","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}