# Probing the Limits of Stylistic Alignment in Vision-Language Models

arXiv: https://arxiv.org/abs/2509.25568 · HTML: https://mlsanity.com/paper/2509.25568/

## 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 20 of its 25 references that are in the map.
- First posted on arXiv 29 Sep 2025.

## Placed in

- Multimodal LLMs / vision-language models / [VLM benchmarks and visual reasoning](https://mlsanity.com/mllm/vlm-benchmarks-and-visual-reasoning/)
- LLM alignment / RLHF / preference optimisation / [Diffusion and multimodal model alignment](https://mlsanity.com/alignment/diffusion-and-multimodal-model-alignment/)

## Builds on (most cited first, 20 of 20)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Prefix-Tuning: Optimizing Continuous Prompts for Generation](https://mlsanity.com/paper/2101.00190/) (2021)
- [Gaussian Error Linear Units (GELUs)](https://mlsanity.com/paper/1606.08415/) (2016)
- [Fine-Tuning Language Models from Human Preferences](https://mlsanity.com/paper/1909.08593/) (2019)
- [SimPO: Simple Preference Optimization with a Reference-Free Reward](https://mlsanity.com/paper/2405.14734/) (2024)
- [Vision-Language Models for Vision Tasks: A Survey](https://mlsanity.com/paper/2304.00685/) (2023)
- [ORPO: Monolithic Preference Optimization without Reference Model](https://mlsanity.com/paper/2403.07691/) (2024)
- [A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applications](https://mlsanity.com/paper/2402.07927/) (2024)
- [Enhancing the Reasoning Ability of Multimodal Large Language Models via Mixed Preference Optimization](https://mlsanity.com/paper/2411.10442/) (2024)
- [Personalized Steering of Large Language Models: Versatile Steering Vectors Through Bi-directional Preference Optimization](https://mlsanity.com/paper/2406.00045/) (2024)
- [Do Androids Laugh at Electric Sheep? Humor "Understanding" Benchmarks from The New Yorker Caption Contest](https://mlsanity.com/paper/2209.06293/) (2022)
- ["Factual" or "Emotional": Stylized Image Captioning with Adaptive Learning and Attention](https://mlsanity.com/paper/1807.03871/) (2018)
- [Painting with Words: Elevating Detailed Image Captioning with Benchmark and Alignment Learning](https://mlsanity.com/paper/2503.07906/) (2025)
- [Altogether: Image Captioning via Re-aligning Alt-text](https://mlsanity.com/paper/2410.17251/) (2024)
- [Massive Supervised Fine-tuning Experiments Reveal How Data, Layer, and Training Factors Shape LLM Alignment Quality](https://mlsanity.com/paper/2506.14681/) (2025)
- [Intuitive Fine-Tuning: Towards Simplifying Alignment into a Single Process](https://mlsanity.com/paper/2405.11870/) (2024)
- [Towards Generating Stylized Image Captions via Adversarial Training](https://mlsanity.com/paper/1908.02943/) (2019)
- [VickreyFeedback: Cost-efficient Data Construction for Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2409.18417/) (2024)

## Abstract (arXiv)

Vision-language models are increasingly used to generate image captions in specific styles, such as humor or romantic. However, these transformer-based models often struggle with this subjective task in a zero-shot setting. While preference data can be used to align them toward a desired style, such data is expensive to acquire, limiting the ability to explore the models' full capabilities. This work addresses this by studying the data efficiency of aligning small vision-language models to humor and romantic styles. This approach helps to define the performance limits of these models and determine how little preference data is needed to achieve stylistic saturation, benchmarking their capabilities and limitations.