SOURCE-LINKED INTELLIGENCE

Lost in Reordering: Structural Sensitivity of Multilingual LLMs under Semantics-Preserving Perturbations

arXiv · AI, language, vision and robotics · article · Sep 3, 2026 · UTC

Large Language Models (LLMs) demonstrate strong multilingual reasoning performance, yet their robustness to semantics-preserving structural variation remains underexplored, particularly for relatively free word-order languages. We investigate the structural sensitivity of multilingual LLMs using two linguistically grounded perturbation settings in Hindi and Malayalam: constrained constituent reordering and active-passive voice transformation. We introduce a benchmark dataset IndicReStruct, with two variants, GSM8K-Reordered and GSM8K-Voice, constructed from GSM8K while preserving semantic mean

Read original source ↗ Open in workspace

recordType

paper

region

Global