Iterative In-Context Learning to Enhance LLMs Abstract Reasoning: The Case-Study of Algebraic Tasks
Proceedings of the 41st ACM/SIGAPP Symposium on Applied Computing (SAC '26), Thessaloniki, Greece, pages 908-916
Presented on
In short
The paper proposes an in-context learning method that iteratively picks few-shot examples to help LLMs solve algebraic expressions under non-standard rules in which the priority of addition and multiplication is swapped. LLMs handled these tasks poorly, but iterative example selection combined with explicit reasoning instructions helped, and some models generalized better when given simpler examples.
Cite
@inproceedings{Fioravanti_2026,
series={SAC ’26},
title={Iterative In-Context Learning to Enhance LLMs Abstract Reasoning: The Case-Study of Algebraic Tasks},
url={http://dx.doi.org/10.1145/3748522.3779772},
DOI={10.1145/3748522.3779772},
booktitle={Proceedings of the 41st ACM/SIGAPP Symposium on Applied Computing},
publisher={ACM},
author={Fioravanti, Stefano and Zavatteri, Matteo and Confalonieri, Roberto and Zeinalipour, Kamyar and Frazzetto, Paolo and Sperduti, Alessandro and Navarin, Nicolò},
year={2026},
month=Mar,
pages={908–916},
collection={SAC ’26}
}