ASyMOB: Algebraic Symbolic Mathematical Operations Benchmark

Michael Shalyt, Rotem Elimelech, Ido Kaminer
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:109814-109835, 2026.

Abstract

Large language models (LLMs) are increasingly applied to symbolic mathematics, yet existing evaluations often conflate pattern memorization with genuine reasoning. To address this gap, we present ASyMOB, a high-resolution dataset of 35,368 validated symbolic math problems spanning integration, limits, differential equations, series, and hypergeometrics. Unlike prior benchmarks, ASyMOB systematically perturbs each seed problem using symbolic, numeric, and equivalence-preserving transformations, enabling a fine-grained assessment of generalization and robustness. Our evaluation reveals three key findings: (1) most models’ performance collapses under minor perturbations, while top systems exhibit an apparent regime shift in robustness; (2) integrated code tools stabilize performance, particularly for weaker models; and (3) we identify examples where Computer Algebra Systems (CAS) fail while LLMs succeed, as well as problems solved only via a hybrid LLM-CAS approach, highlighting a promising integration frontier. ASyMOB serves as a principled diagnostic tool for measuring and accelerating progress toward building verifiable, trustworthy AI for scientific discovery.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-shalyt26a, title = {{AS}y{MOB}: Algebraic Symbolic Mathematical Operations Benchmark}, author = {Shalyt, Michael and Elimelech, Rotem and Kaminer, Ido}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {109814--109835}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/shalyt26a/shalyt26a.pdf}, url = {https://proceedings.mlr.press/v306/shalyt26a.html}, abstract = {Large language models (LLMs) are increasingly applied to symbolic mathematics, yet existing evaluations often conflate pattern memorization with genuine reasoning. To address this gap, we present ASyMOB, a high-resolution dataset of 35,368 validated symbolic math problems spanning integration, limits, differential equations, series, and hypergeometrics. Unlike prior benchmarks, ASyMOB systematically perturbs each seed problem using symbolic, numeric, and equivalence-preserving transformations, enabling a fine-grained assessment of generalization and robustness. Our evaluation reveals three key findings: (1) most models’ performance collapses under minor perturbations, while top systems exhibit an apparent regime shift in robustness; (2) integrated code tools stabilize performance, particularly for weaker models; and (3) we identify examples where Computer Algebra Systems (CAS) fail while LLMs succeed, as well as problems solved only via a hybrid LLM-CAS approach, highlighting a promising integration frontier. ASyMOB serves as a principled diagnostic tool for measuring and accelerating progress toward building verifiable, trustworthy AI for scientific discovery.} }
Endnote
%0 Conference Paper %T ASyMOB: Algebraic Symbolic Mathematical Operations Benchmark %A Michael Shalyt %A Rotem Elimelech %A Ido Kaminer %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-shalyt26a %I PMLR %P 109814--109835 %U https://proceedings.mlr.press/v306/shalyt26a.html %V 306 %X Large language models (LLMs) are increasingly applied to symbolic mathematics, yet existing evaluations often conflate pattern memorization with genuine reasoning. To address this gap, we present ASyMOB, a high-resolution dataset of 35,368 validated symbolic math problems spanning integration, limits, differential equations, series, and hypergeometrics. Unlike prior benchmarks, ASyMOB systematically perturbs each seed problem using symbolic, numeric, and equivalence-preserving transformations, enabling a fine-grained assessment of generalization and robustness. Our evaluation reveals three key findings: (1) most models’ performance collapses under minor perturbations, while top systems exhibit an apparent regime shift in robustness; (2) integrated code tools stabilize performance, particularly for weaker models; and (3) we identify examples where Computer Algebra Systems (CAS) fail while LLMs succeed, as well as problems solved only via a hybrid LLM-CAS approach, highlighting a promising integration frontier. ASyMOB serves as a principled diagnostic tool for measuring and accelerating progress toward building verifiable, trustworthy AI for scientific discovery.
APA
Shalyt, M., Elimelech, R. & Kaminer, I.. (2026). ASyMOB: Algebraic Symbolic Mathematical Operations Benchmark. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:109814-109835 Available from https://proceedings.mlr.press/v306/shalyt26a.html.

Related Material