SOURCE-LINKED INTELLIGENCE

Cross-Lingual Alignment Without Joint Training: Do Monolingual Language Models Converge on Universal Representations?

arXiv · AI, language, vision and robotics · article · Aug 27, 2026 · UTC

Cross-lingual alignment in multilingual language models is typically attributed to joint training: shared parameters, mixed-language batches, or explicit alignment objectives. We ask whether monolingual models trained on non-parallel data learn alignable representations without joint training. By testing on strictly monolingual language models, such as the Goldfish model families and independently developed models from different research labs, we find three results. Correlation: these models develop alignable representational geometry across layers, with alignment strengthening as data scale,

Read original source ↗ Open in workspace

recordType

paper

region

Global