cs.CLJun 8, 2026

SEF-CLGC at SemEval-2026 Task 11: Logical Notation Impact on Language Model Performance

Authors: Hanna Abi Akl, Fabien Gandon, Catherine Faron, Pierre Monnin

Organizations: Université Côte d’Azur, Inria, CNRS, I3S, Sophia Antipolis, France · Data ScienceTech Institute, Paris, France

Abstract

This paper revisits our pipeline called Syllogistic Evaluation Framework-Common Logic Grammar Construction (SEF-CLGC). We combine formal logical notations with Small Language Models (SLMs) to evaluate reasoning performance on the SemEval-2026 Task 11 Subtask 1: Disentangling Content and Formal Reasoning in Large Language Models. Our experiments show that by relying solely on SLMs, trained on a combination of natural and symbolic languages, our best model achieves a content score of 27.80% on the task while significantly lowering the content bias in reasoning.

Explore similar work

CardsList
  1. Are you Talking Logic to Me? Assessing Language Models Syllogistic Reasoning Capabilities

    Jul 22, 2026Hanna Abi Akl, Fabien Gandon, Catherine Faron +1