# Debiasing Multimodal Large Language Models via Noise-Aware Preference Optimization

arXiv: https://arxiv.org/abs/2503.17928 · HTML: https://mlsanity.com/paper/2503.17928/

## Key facts

- Cited by 14 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 56 of its 66 references that are in the map.
- First posted on arXiv 23 Mar 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, 30 of 56)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models](https://mlsanity.com/paper/2304.10592/) (2023)
- [A Survey on LLM-as-a-Judge](https://mlsanity.com/paper/2411.15594/) (2024)
- [Instruction-Following Evaluation for Large Language Models](https://mlsanity.com/paper/2311.07911/) (2023)
- [mPLUG-Owl: Modularization Empowers Large Language Models with Multimodality](https://mlsanity.com/paper/2304.14178/) (2023)
- [Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labels](https://mlsanity.com/paper/1805.07836/) (2018)
- [Aligning Large Multimodal Models with Factually Augmented RLHF](https://mlsanity.com/paper/2309.14525/) (2023)
- [Mitigating Object Hallucinations in Large Vision-Language Models through Visual Contrastive Decoding](https://mlsanity.com/paper/2311.16922/) (2023)
- [RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment](https://mlsanity.com/paper/2304.06767/) (2023)
- [Object Hallucination in Image Captioning](https://mlsanity.com/paper/1809.02156/) (2018)
- [ORPO: Monolithic Preference Optimization without Reference Model](https://mlsanity.com/paper/2403.07691/) (2024)
- [From Generation to Judgment: Opportunities and Challenges of LLM-as-a-judge](https://mlsanity.com/paper/2411.16594/) (2024)
- [OPERA: Alleviating Hallucination in Multi-Modal Large Language Models via Over-Trust Penalty and Retrospection-Allocation](https://mlsanity.com/paper/2311.17911/) (2023)
- [RRHF: Rank Responses to Align Language Models with Human Feedback without tears](https://mlsanity.com/paper/2304.05302/) (2023)
- [Large Language Models Struggle to Learn Long-Tail Knowledge](https://mlsanity.com/paper/2211.08411/) (2022)
- [RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human Feedback](https://mlsanity.com/paper/2312.00849/) (2023)
- [SLiC-HF: Sequence Likelihood Calibration with Human Feedback](https://mlsanity.com/paper/2305.10425/) (2023)
- [Statistical Rejection Sampling Improves Preference Optimization](https://mlsanity.com/paper/2309.06657/) (2023)
- [AMBER: An LLM-free Multi-dimensional Benchmark for MLLMs Hallucination Evaluation](https://mlsanity.com/paper/2311.07397/) (2023)
- [Counterfactual VQA: A Cause-Effect Look at Language Bias](https://mlsanity.com/paper/2006.04315/) (2020)
- [Self-Play Preference Optimization for Language Model Alignment](https://mlsanity.com/paper/2405.00675/) (2024)
- [Beyond Hallucinations: Enhancing LVLMs through Hallucination-Aware Direct Preference Optimization](https://mlsanity.com/paper/2311.16839/) (2023)

## Built on by (most cited first, 14 of 14)

- [MLLMs are Deeply Affected by Modality Bias](https://mlsanity.com/paper/2505.18657/) (2025)
- [Multi-speaker Attention Alignment for Multimodal Social Interaction](https://mlsanity.com/paper/2511.17952/) (2025)
- [Residual Decoding: Mitigating Hallucinations in Large Vision-Language Models via History-Aware Residual Guidance](https://mlsanity.com/paper/2602.01047/) (2026)
- [Investigating VLM Hallucination from a Cognitive Psychology Perspective: A First Step Toward Interpretation with Intriguing Observations](https://mlsanity.com/paper/2507.03123/) (2025)
- [AMPS: Adaptive Modality Preference Steering via Functional Entropy](https://mlsanity.com/paper/2602.12533/) (2026)
- [Unleashing the Intrinsic Visual Representation Capability of Multimodal Large Language Models](https://mlsanity.com/paper/2512.06281/) (2025)
- [CSR-Bench: A Benchmark for Evaluating the Cross-modal Safety and Reliability of MLLMs](https://mlsanity.com/paper/2602.03263/) (2026)
- [Beyond Perceptual Shortcuts: Causal-Inspired Debiasing Optimization for Generalizable Video Reasoning in Lightweight MLLMs](https://mlsanity.com/paper/2605.01324/) (2026)
- [LaViPlan : Language-Guided Visual Path Planning with RLVR](https://mlsanity.com/paper/2507.12911/) (2025)
- [Bias Mitigation Agent: Optimizing Source Selection for Fair and Balanced Knowledge Retrieval](https://mlsanity.com/paper/2508.18724/) (2025)
- [QIRL: Optimized Question-Image Relation Learning for Bias-Robust Visual Question Answering](https://mlsanity.com/paper/2504.03337/) (2025)
- [Thought-For-Food: Reasoning Chain Induced Food Visual Question Answering](https://mlsanity.com/paper/2511.01213/) (2025)
- [EgoArgus: Benchmarking VLMs as Situational Assistants for Modality-Grounded User Supports](https://mlsanity.com/paper/2608.25561/) (2026)
- [Sensory Precision Inference for Multimodal Arbitration under Uncertainty](https://mlsanity.com/paper/2609.15065/) (2026)

## Abstract (arXiv)

Multimodal Large Language Models excel in various tasks, yet often struggle with modality bias, where the model tends to rely heavily on a single modality and overlook critical information in other modalities, which leads to incorrect focus and generating irrelevant responses. In this paper, we propose using the paradigm of preference optimization to solve the modality bias problem, including RLAIFVBias, a debiased preference optimization dataset, and a Noise Aware Preference Optimization algorithm. Specifically, we first construct the dataset by introducing perturbations to reduce the informational content of certain modalities, compelling the model to rely on a specific modality when generating negative responses. To address the inevitable noise in automatically constructed data, we combine the noise robust Mean Absolute Error with the Binary Cross Entropy in Direct Preference Optimization by a negative Box Cox transformation, and dynamically adjust the algorithm noise robustness based on the evaluated noise levels in the data. Extensive experiments validate our approach, demonstrating not only its effectiveness in mitigating modality bias but also its significant role in minimizing hallucinations.