# VisualDeltas: Learning Preferences from Visual Quality Perturbations

arXiv: https://arxiv.org/abs/2603.07272 · HTML: https://mlsanity.com/paper/2603.07272/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 19 of its 26 references that are in the map.
- First posted on arXiv 7 Mar 2026.

## Builds on (most cited first, 19 of 19)

- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering](https://mlsanity.com/paper/1612.00837/) (2016)
- [Self-Rewarding Language Models](https://mlsanity.com/paper/2401.10020/) (2024)
- [Measuring Multimodal Mathematical Reasoning with MATH-Vision Dataset](https://mlsanity.com/paper/2402.14804/) (2024)
- [Compositional Semantic Parsing on Semi-Structured Tables](https://mlsanity.com/paper/1508.00305/) (2015)
- [Direct Preference Optimization of Video Large Multimodal Models from Language Model Reward](https://mlsanity.com/paper/2404.01258/) (2024)
- [MM-RLHF: The Next Step Forward in Multimodal LLM Alignment](https://mlsanity.com/paper/2502.10391/) (2025)
- [Relative Preference Optimization: Enhancing LLM Alignment through Contrasting Responses across Identical and Diverse Prompts](https://mlsanity.com/paper/2402.10958/) (2024)
- [The Delta Learning Hypothesis: Preference Tuning on Weak Data can Yield Strong Gains](https://mlsanity.com/paper/2507.06187/) (2025)
- [Modality-Fair Preference Optimization for Trustworthy MLLM Alignment](https://mlsanity.com/paper/2410.15334/) (2024)
- [MallowsPO: Fine-Tune Your LLM with Preference Dispersions](https://mlsanity.com/paper/2405.14953/) (2024)
- [Mitigating Hallucinations in Large Vision-Language Models via Entity-Centric Multimodal Preference Optimization](https://mlsanity.com/paper/2506.04039/) (2025)
- [Robust-R1: Degradation-Aware Reasoning for Robust Visual Understanding](https://mlsanity.com/paper/2512.17532/) (2025)
- [AdaViP: Aligning Multi-modal LLMs via Adaptive Vision-enhanced Preference Optimization](https://mlsanity.com/paper/2504.15619/) (2025)
- [MutualVPR: A Mutual Learning Framework for Resolving Supervision Inconsistencies via Adaptive Clustering](https://mlsanity.com/paper/2412.09199/) (2024)
- [LPOI: Listwise Preference Optimization for Vision Language Models](https://mlsanity.com/paper/2505.21061/) (2025)
- [M3PO: Multimodal-Model-Guided Preference Optimization for Visual Instruction Following](https://mlsanity.com/paper/2508.12458/) (2025)
- [Robust Diagram Reasoning: A Framework for Enhancing LVLM Performance on Visually Perturbed Scientific Diagrams](https://mlsanity.com/paper/2508.16972/) (2025)

## Abstract (arXiv)

We present VisualDeltas, a lightweight preference-learning framework that extracts supervision from visual quality variations in multimodal data. By leveraging the systematic impact of image quality on visual perception and reasoning, VisualDeltas induces informative preference signals without relying on human annotations or external teachers. The framework supports both label-free and label-based regimes, enabling flexible use of available supervision when present. Across diverse multimodal benchmarks and model scales, VisualDeltas consistently outperforms rejection-sampling fine-tuning and improves generalization, and extends naturally to a range of visual degradations.