# Not Truly Multilingual: Script Consistency as a Missing Dimension in VLM Evaluation

arXiv: https://arxiv.org/abs/2606.17188 · HTML: https://mlsanity.com/paper/2606.17188/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in VLM benchmarks and visual reasoning, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 15 of its 20 references that are in the map.
- First posted on arXiv 15 Jun 2026.

## Placed in

- Multimodal LLMs / vision-language models / [VLM benchmarks and visual reasoning](https://mlsanity.com/mllm/vlm-benchmarks-and-visual-reasoning/)

## Builds on (most cited first, 15 of 15)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond](https://mlsanity.com/paper/2308.12966/) (2023)
- [How Good is Your Tokenizer? On the Monolingual Performance of Multilingual Language Models](https://mlsanity.com/paper/2012.15613/) (2020)
- [Visually Grounded Reasoning across Languages and Cultures](https://mlsanity.com/paper/2109.13238/) (2021)
- [UNKs Everywhere: Adapting Multilingual Language Models to New Scripts](https://mlsanity.com/paper/2012.15562/) (2020)
- [Crossmodal-3600: A Massively Multilingual Multimodal Evaluation Dataset](https://mlsanity.com/paper/2205.12522/) (2022)
- [Broaden the Vision: Geo-Diverse Visual Commonsense Reasoning](https://mlsanity.com/paper/2109.06860/) (2021)
- [All Languages Matter: Evaluating LMMs on Culturally Diverse 100 Languages](https://mlsanity.com/paper/2411.16508/) (2024)
- [IGLUE: A Benchmark for Transfer Learning across Modalities, Tasks, and Languages](https://mlsanity.com/paper/2201.11732/) (2022)
- [PALO: A Polyglot Large Multimodal Model for 5B People](https://mlsanity.com/paper/2402.14818/) (2024)
- [IndicVisionBench: Benchmarking Cultural and Multilingual Understanding in VLMs](https://mlsanity.com/paper/2511.04727/) (2025)
- [MVL-SIB: A Massively Multilingual Vision-Language Benchmark for Cross-Modal Topical Matching](https://mlsanity.com/paper/2502.12852/) (2025)
- [Script Gap: Evaluating LLM Triage on Indian Languages in Native vs Romanized Scripts in a Real World Setting](https://mlsanity.com/paper/2512.10780/) (2025)
- [CORI: CJKV Benchmark with Romanization Integration -- A step towards Cross-lingual Transfer Beyond Textual Scripts](https://mlsanity.com/paper/2404.12618/) (2024)
- [BLEnD-Vis: Benchmarking Multimodal Cultural Understanding in Vision Language Models](https://mlsanity.com/paper/2510.11178/) (2025)

## Abstract (arXiv)

Current multilingual evaluations for Vision-Language Models (VLMs) assume a one-to-one mapping between language and orthography, overlooking billions of users of multi-script languages. We introduce PuMVR (Punjabi Multimodal Visual Reasoning), a benchmark of 1,000 strictly parallel image-text instances across Punjabi's three active scripts: Gurmukhi, Shahmukhi, and Roman. Evaluating 10 state-of-the-art VLMs, we expose a substantial and systematic Script Gap. Models frequently solve visual tasks in one script while failing identical tasks in another, with accuracy deltas reaching 16%. Crucially, visual input boosts absolute performance uniformly yet does not close the orthographic gap. Furthermore, cross-script in-context transfer is highly brittle, exposing script-locked knowledge representation. Supported by McNemar tests across all script pairs, our findings demonstrate that current "multilingual" VLMs are not truly multi-script. We propose the Script Consistency Rate (SCR), which falls as low as 24.8% on our benchmark, as a mandatory metric for script-agnostic evaluation to ensure equitable AI access. Data and code are available at: https://github.com/prabhjotschugh/Not-Truly-Multilingual-PuMVR.