Abstract:This survey examines multilingual vision-language models that process text and images across languages. We review 33 models and 23 benchmarks, spanning encoder-only and generative architectures, and identify a key tension between language neutrality (consistent cross-lingual representations) and cultural awareness (adaptation to cultural contexts). Current training methods favor neutrality through contrastive learning, while cultural awareness depends on diverse data. Two-thirds of evaluation benchmarks use translation-based approaches prioritizing semantic consistency, though recent work incorporates culturally grounded content. We find discrepancies in cross-lingual capabilities and gaps between training objectives and evaluation goals.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2509.22123 [cs.CL] |
| (or arXiv:2509.22123v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2509.22123 arXiv-issued DOI via DataCite |
Submission history
From: Andrei-Alexandru Manea [view email]
[v1]
Fri, 26 Sep 2025 09:46:13 UTC (9,186 KB)
[v2]
Wed, 13 May 2026 12:28:55 UTC (9,840 KB)