# Parameter Efficient Multimodal Instruction Tuning for Romanian Vision Language Models

Source: [Parameter Efficient Multimodal Instruction Tuning for Romanian Vision Language Models](<https://www.sota2.com/research/paper/69438a63423ea4226572927d>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-07-08T05:00:20.838Z

Authors: George-Andrei Dima, R\\u{a}zvan-Alexandru Sm\\u{a}du, Dumitru-Clementin Cercel

Published: 2025-12-16T21:36:28.000Z

## Abstract

Focusing on low-resource languages is an essential step toward democratizing generative AI. In this work, we contribute to reducing the multimodal NLP resource gap for Romanian. We translate the widely known Flickr30K dataset into Romanian and further extend it for visual question answering by leveraging open-source LLMs. We demonstrate the usefulness of our datasets by fine-tuning open-source VLMs on Romanian visual question answering. We select VLMs from three widely used model families: LLaMA 3.2, LLaVA 1.6, and Qwen2. For fine-tuning, we employ the parameter-efficient LoRA method. Our models show improved Romanian capabilities in visual QA, as well as on tasks they were not trained on, such as Romanian image description generation. The seven-billion-parameter Qwen2-VL-RoVQA obtains top scores on both tasks, with improvements of +2.29% and +4.45% in BERTScore F1 on VQA and captioning, respectively, over its original version. Finally, the models show substantial reductions in grammatical errors compared to their original forms, indicating improvements not only in language understanding but also in Romanian fluency.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Visual Question Answering on Flickr30k RoQA (test)](<https://www.sota2.com/research/sota/visual-question-answering-on-flickr30k-roqa-test.md>): Visual Question Answering; Flickr30k RoQA (test); BERTScore P: 80.43.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2512.14926>)

[PDF](<https://arxiv.org/pdf/2512.14926.pdf>)