A Pilot Study on Curator-Guided Multilingual Art Description for Blind and Low-Vision Audiences with Small Vision-Language Models
On the map: Multimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Abstract (arXiv)
Blind and low-vision (BLV) audiences remain underserved by visual art descriptions, particularly across languages and in museum settings where privacy and intellectual-property constraints may favour small on-premise vision-language models (VLMs). This pilot study investigates curator-guided multilingual art description with Qwen2.5-VL-3B-Instruct for German, Romanian, and Serbian. We construct a parallel BLV-oriented caption corpus from artwork images and metadata, and compare language-specific LoRA adapters with a single multilingual adapter under a fixed backbone and training budget. Evaluation combines automatic lexical and embedding-based metrics with an LLM-as-Judge protocol calibrated against a small Romanian BLV pilot study. Under our pilot setup, language-specific adapters show more stable controllability and visually grounded description quality for Romanian and Serbian, while multilingual adaptation remains competitive in German. We frame these findings as deployment-oriented evidence for small on-premise VLMs, and highlight the need for larger BLV user studies and broader language coverage before drawing general conclusions about multilingual accessibility.
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 17 of its 33 references that are in the map. First posted on arXiv 29 May 2026.
Builds on
17 of 33 references.
References · 17
- 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
- 2021On the Opportunities and Risks of Foundation Modelscited by 4,033
- 2023G-Eval: NLG Evaluation using GPT-4 with Better Human Alignmentcited by 2,276
- 2024Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preferencecited by 1,184
- 2024Length-Controlled AlpacaEval: A Simple Way to Debias Automatic Evaluatorscited by 897
- 2021The FLORES-101 Evaluation Benchmark for Low-Resource and Multilingual Machine Translationcited by 560
- 2023Is ChatGPT a Good NLG Evaluator? A Preliminary Studycited by 422
- 2023Large Language Models Are State-of-the-Art Evaluators of Translation Qualitycited by 420
- 2017Why We Need New Evaluation Metrics for NLGcited by 233
- 2021CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representationcited by 232
- 2020Captioning Images Taken by People Who Are Blindcited by 110
- 2021ArtEmis: Affective Language for Visual Artcited by 104
5 more
- 2024Misfitting With AI: How Blind People Verify and Contest AI Errorscited by 15
- 2024Large language models for mental healthcited by 9
- 2024Computer Audition: From Task-Specific Machine Learning to Foundation Modelscited by 8
- 2024Affective Computing Has Changed: The Foundation Model Disruptioncited by 7
- 2025Reading Smiles: Proxy Bias in Foundation Models for Facial Emotion Recognitioncited by 1
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2025A Multimodal Recaptioning Framework to Account for Perceptual Diversity Across Languages in Vision-Language Modeling
- 2024Vision-Language Models under Cultural and Inclusive Considerations
- 2026"Inctelegi Romanecste?'' A Recipe for Romanian Vision-Language Models
- 2024Quantifying the Gaps Between Translation and Native Perception in Training for Multimodal, Multilingual Retrieval
- 2025Behind Maya: Building a Multilingual Vision Language Model
- 2024Towards Artwork Explanation in Large-scale Vision Language Models
- 2025Rethinking Multilingual Vision-Language Translation: Dataset, Evaluation, and Adaptation
- 2026Annotation-Efficient Vision-Language Model Adaptation to the Polish Language Using the LLaVA Framework