The Validation Gap: A Mechanistic Analysis of How Language Models Compute Arithmetic but Fail to Validate It
Published in ACL Main 2025, 2025
Recommended citation: Leonardo Bertolazzi, Philipp Mondorf, Barbara Plank, and Raffaella Bernardi. 2025. The Validation Gap: A Mechanistic Analysis of How Language Models Compute Arithmetic but Fail to Validate It. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 29387–29424, Suzhou, China. Association for Computational Linguistics.
Download Paper