A Closer Look at Tool-based Logical Reasoning with LLMs: The Choice of Tool Matters

Long Hei Matthew Lam, Ramya Keerthy Thatikonda, Ehsan Shareghi


Abstract
The emergence of Large Language Models (LLMs) has demonstrated promising progress in solving logical reasoning tasks effectively. Several recent approaches have proposed to change the role of the LLM from the reasoner into a translator between natural language statements and symbolic representations which are then sent to external symbolic solvers to resolve. This paradigm has established the current state-of-the-art result in logical reasoning (i.e., deductive reasoning). However, it remains unclear whether the variance in performance of these approaches stems from the methodologies employed or the specific symbolic solvers utilized. There is a lack of consistent comparison between symbolic solvers and how they influence the overall reported performance. This is important, as each symbolic solver also has its own input symbolic language, presenting varying degrees of challenge in the translation process. To address this gap, we perform experiments on 3 deductive reasoning benchmarks with LLMs augmented with widely used symbolic solvers: Z3, Pyke, and Prover9. The tool-executable rates of symbolic translation generated by different LLMs exhibit a near 50% performance variation. This highlights a significant difference in performance rooted in very basic choices of tools. The almost linear correlation between the executable rate of translations and the accuracy of the outcomes from Prover9 highlight a strong alignment between LLMs ability to translate into Prover9 symbolic language, and the correctness of those translations.
Anthology ID:
2024.alta-1.4
Volume:
Proceedings of the 22nd Annual Workshop of the Australasian Language Technology Association
Month:
December
Year:
2024
Address:
Canberra, Australia
Editors:
Tim Baldwin, Sergio José Rodríguez Méndez, Nicholas Kuo
Venue:
ALTA
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
41–63
Language:
URL:
https://aclanthology.org/2024.alta-1.4/
DOI:
Bibkey:
Cite (ACL):
Long Hei Matthew Lam, Ramya Keerthy Thatikonda, and Ehsan Shareghi. 2024. A Closer Look at Tool-based Logical Reasoning with LLMs: The Choice of Tool Matters. In Proceedings of the 22nd Annual Workshop of the Australasian Language Technology Association, pages 41–63, Canberra, Australia. Association for Computational Linguistics.
Cite (Informal):
A Closer Look at Tool-based Logical Reasoning with LLMs: The Choice of Tool Matters (Matthew Lam et al., ALTA 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.alta-1.4.pdf