# M3PO: Multimodal-Model-Guided Preference Optimization for Visual Instruction Following

arXiv: https://arxiv.org/abs/2508.12458 · HTML: https://mlsanity.com/paper/2508.12458/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in MLLM training and visual instruction tuning, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 23 of its 35 references that are in the map.
- First posted on arXiv 17 Aug 2025.

## Placed in

- Multimodal LLMs / vision-language models / [MLLM training and visual instruction tuning](https://mlsanity.com/mllm/mllm-training-and-visual-instruction-tuning/)
- LLM alignment / RLHF / preference optimisation / [Diffusion and multimodal model alignment](https://mlsanity.com/alignment/diffusion-and-multimodal-model-alignment/)

## Builds on (most cited first, 23 of 23)

- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://mlsanity.com/paper/2305.06500/) (2023)
- [Video-LLaVA: Learning United Visual Representation by Alignment Before Projection](https://mlsanity.com/paper/2311.10122/) (2023)
- [The Dawn of LMMs: Preliminary Explorations with GPT-4V(ision)](https://mlsanity.com/paper/2309.17421/) (2023)
- [Is DPO Superior to PPO for LLM Alignment? A Comprehensive Study](https://mlsanity.com/paper/2404.10719/) (2024)
- [LVLM-eHub: A Comprehensive Evaluation Benchmark for Large Vision-Language Models](https://mlsanity.com/paper/2306.09265/) (2023)
- [An Introduction to Vision-Language Modeling](https://mlsanity.com/paper/2405.17247/) (2024)
- [Rewards-in-Context: Multi-objective Alignment of Foundation Models with Dynamic Preference Adjustment](https://mlsanity.com/paper/2402.10207/) (2024)
- [Personalizing Reinforcement Learning from Human Feedback with Variational Preference Learning](https://mlsanity.com/paper/2408.10075/) (2024)
- [Visual In-Context Learning for Large Vision-Language Models](https://mlsanity.com/paper/2402.11574/) (2024)
- [Vision-Language Models for Medical Report Generation and Visual Question Answering: A Review](https://mlsanity.com/paper/2403.02469/) (2024)
- [Thread of Thought Unraveling Chaotic Contexts](https://mlsanity.com/paper/2311.08734/) (2023)
- [MAM: Modular Multi-Agent Framework for Multi-Modal Medical Diagnosis via Role-Specialized Collaboration](https://mlsanity.com/paper/2506.19835/) (2025)
- [Visual-Text Cross Alignment: Refining the Similarity Score in Vision-Language Models](https://mlsanity.com/paper/2406.02915/) (2024)
- [Draw ALL Your Imagine: A Holistic Benchmark and Agent Framework for Complex Instruction-based Image Generation](https://mlsanity.com/paper/2505.24787/) (2025)
- [Aligning Language Models with Human Preferences via a Bayesian Approach](https://mlsanity.com/paper/2310.05782/) (2023)
- [Improving Cross-modal Alignment for Text-Guided Image Inpainting](https://mlsanity.com/paper/2301.11362/) (2023)
- [Cross-Modal Safety Mechanism Transfer in Large Vision-Language Models](https://mlsanity.com/paper/2410.12662/) (2024)
- [Preference Alignment with Flow Matching](https://mlsanity.com/paper/2405.19806/) (2024)
- [A Preference-driven Paradigm for Enhanced Translation with Large Language Models](https://mlsanity.com/paper/2404.11288/) (2024)
- [Activating Distributed Visual Region within LLMs for Efficient and Effective Vision-Language Training and Inference](https://mlsanity.com/paper/2412.12785/) (2024)

## Built on by (most cited first, 1 of 1)

- [VisualDeltas: Learning Preferences from Visual Quality Perturbations](https://mlsanity.com/paper/2603.07272/) (2026)

## Abstract (arXiv)

Large Vision-Language Models (LVLMs) hold immense potential for complex multimodal instruction following, yet their development is often hindered by the high cost and inconsistency of human annotation required for effective fine-tuning and preference alignment. Traditional supervised fine-tuning (SFT) and existing preference optimization methods like RLHF and DPO frequently struggle to efficiently leverage the model's own generation space to identify highly informative "hard negative" samples. To address these challenges, we propose Multimodal-Model-Guided Preference Optimization (M3PO), a novel and data-efficient method designed to enhance LVLMs' capabilities in visual instruction following. M3PO intelligently selects the most "learning-valuable" preference sample pairs from a diverse pool of LVLM-generated candidates. This selection is driven by a sophisticated mechanism that integrates two crucial signals: a Multimodal Alignment Score (MAS) to assess external quality and the model's Self-Consistency / Confidence (log-probability) to gauge internal belief. These are combined into a novel M3P-Score, which specifically identifies preferred responses and challenging dispreferred responses that the model might confidently generate despite being incorrect. These high-quality preference pairs are then used for efficient Direct Preference Optimization (DPO) fine-tuning on base LVLMs like LLaVA-1.5 (7B/13B) using LoRA. Our extensive experiments demonstrate that M3PO consistently outperforms strong baselines, including SFT, simulated RLHF, vanilla DPO, and RM-DPO, across a comprehensive suite of multimodal instruction following benchmarks (MME-Bench, POPE, IFT, Human Pref. Score).