mlsanityData to 14 Oct 2026

Mitigating Hallucination Through Theory-Consistent Symmetric Multimodal Preference Optimization

arXiv 2506.11712 ↗First posted 13 Jun 2025cs.AIcited by 14

On the map: Multimodal LLMs / vision-language models → Hallucination mitigation in LVLMs

also in 1 more
Abstract (arXiv)

Direct Preference Optimization (DPO) has emerged as an effective approach for mitigating hallucination in Multimodal Large Language Models (MLLMs). Although existing methods have achieved significant progress by utilizing vision-oriented contrastive objectives for enhancing MLLMs' attention to visual inputs and hence reducing hallucination, they suffer from non-rigorous optimization objective function and indirect preference supervision. To address these limitations, we propose a Symmetric Multimodal Preference Optimization (SymMPO), which conducts symmetric preference learning with direct preference supervision (i.e., response pairs) for visual understanding enhancement, while maintaining rigorous theoretical alignment with standard DPO. In addition to conventional ordinal preference learning, SymMPO introduces a preference margin consistency loss to quantitatively regulate the preference gap between symmetric preference pairs. Comprehensive evaluation across five benchmarks demonstrate SymMPO's superior performance, validating its effectiveness in hallucination mitigation of MLLMs.

Key facts. Cited by 14 papers in the map (data through 14 Oct 2026). Placed in Hallucination mitigation in LVLMs, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 42 of its 50 references that are in the map. First posted on arXiv 13 Jun 2025.

Builds on

42 of 50 references.

Closest ancestors · 12

Same direction, newest first

  1. 2025CHiP: Cross-modal Hierarchical Direct Preference Optimization for Multimodal LLMs28 Jan
  2. 2025Mitigating Hallucinations in Large Vision-Language Models via DPO: On-Policy Data Hold the Key16 Jan
  3. 2024V-DPO: Mitigating Hallucination in Large Vision Language Models via Vision-Guided Direct Preference Optimization5 Nov
  4. 2024CLIP-DPO: Vision-Language Models as a Source of Preference for Fixing Hallucinations in LVLMs19 Aug
  5. 2024Automated Multi-level Preference for MLLMs18 May
  6. 2024Hallucination of Multimodal Large Language Models: A Survey29 Apr
  7. 2024Detecting and Mitigating Hallucination in Large Vision Language Models via Fine-Grained AI Feedback22 Apr
  8. 2023RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human Feedback1 Dec
  9. 2023Beyond Hallucinations: Enhancing LVLMs through Hallucination-Aware Direct Preference Optimization28 Nov
  10. 2023AMBER: An LLM-free Multi-dimensional Benchmark for MLLMs Hallucination Evaluation13 Nov
  11. 2023HallusionBench: An Advanced Diagnostic Suite for Entangled Language Hallucination and Visual Illusion in Large Vision-Language Models23 Oct
  12. 2023Detecting and Preventing Hallucinations in Large Vision Language Models11 Aug

Other references · 30

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2023GPT-4 Technical Reportcited by 18,396
  4. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  5. 2022Training language models to follow instructions with human feedbackcited by 15,971
  6. 2017Proximal Policy Optimization Algorithmscited by 15,215
  7. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  8. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 8,065
  9. 2023Visual Instruction Tuningcited by 7,869
  10. 2025Qwen2.5-VL Technical Reportcited by 5,587
  11. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,234
  12. 2023Qwen Technical Reportcited by 3,344
18 more
  1. 2016Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotationscited by 2,656
  2. 2020Learning to summarize from human feedbackcited by 2,588
  3. 2016Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answeringcited by 2,516
  4. 2019Towards VQA Models That Can Readcited by 1,718
  5. 2022RePaint: Inpainting using Denoising Diffusion Probabilistic Modelscited by 1,430
  6. 2024Are We on the Right Way for Evaluating Large Vision-Language Models?cited by 914
  7. 2023Aligning Large Multimodal Models with Factually Augmented RLHFcited by 712
  8. 2018Object Hallucination in Image Captioningcited by 669
  9. 2024Enhancing the Reasoning Ability of Multimodal Large Language Models via Mixed Preference Optimizationcited by 239
  10. 2024Aligning Modalities in Vision Large Language Models via Preference Fine-tuningcited by 198
  11. 2020Improved Contrastive Divergence Training of Energy Based Modelscited by 130
  12. 2024Unpacking DPO and PPO: Disentangling Best Practices for Learning from Preference Feedbackcited by 107
  13. 2024mDPO: Conditional Preference Optimization for Multimodal Large Language Modelscited by 92
  14. 2024Strengthening Multimodal Large Language Model with Bootstrapped Preference Optimizationcited by 75
  15. 2024MIA-DPO: Multi-Image Augmented Direct Preference Optimization For Large Vision-Language Modelscited by 34
  16. 2024Modality-Fair Preference Optimization for Trustworthy MLLM Alignmentcited by 26
  17. 2023Reformulating Vision-Language Foundation Models and Datasets Towards Universal Multimodal Assistantscited by 25
  18. 2024SymDPO: Boosting In-Context Learning of Large Multimodal Models with Symbol Demonstration Direct Preference Optimizationcited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 4; 2026: 10.

14 papers.

Sort:

By area:

  1. 2025Automating Expert-Level Medical Reasoning Evaluation of Large Language Modelscited by 22
  2. 2025Tell Model Where to Look: Mitigating Hallucinations in MLLMs by Vision-Guided Attentioncited by 9
  3. 2025MITS: Enhanced Tree Search Reasoning for LLMs via Pointwise Mutual Informationcited by 8
  4. 2026Visual Preference Optimization with Rubric Rewardscited by 5
  5. 2026Online Skill Learning for Web Agents via State-Grounded Dynamic Retrievalcited by 3
  6. 2026GraphVerse: A Comprehensive Visual Graph Reasoning Benchmark for Multimodal Large Language Modelscited by 2
  7. 2025RSTR: Reducing SpatioTemporal Redundancy in Diffusion Transformerscited by 1
  8. 2026Palette: A Modular, Controllable, and Efficient Framework for On-demand Authorized Safety Alignment Relaxation in LLMscited by 1
  9. 2026Clearer Sight, Fewer Lies: Oriented Pickup Preference Optimization for Multimodal Hallucination Mitigationcited by 1
  10. 2026IRIS: Implicit Reward-Guided Internal Sifting for Mitigating Multimodal Hallucinationcited by 0
  11. 2026Learning from Fine-Grained Visual Discrepancies: Mitigating Multimodal Hallucinations via In-Context Visual Contrastive Optimizationcited by 0
  12. 2026Vision-driven Preference Synthesis for Mitigating Hallucinations in VLMscited by 0
2 more
  1. 2026PEA-DPO: Perception-Enhanced Alignment Direct Preference Optimization for MLLMs Alignmentcited by 0
  2. 2026Beyond Saying Less: Fine-Grained Alignment for Informative and Faithful Vision-Language Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Mitigating Hallucination in Multimodal Large Language Model via Hallucination-targeted Direct Preference Optimization
  2. 2025TARS: MinMax Token-Adaptive Preference Strategy for Hallucination Reduction in MLLMs
  3. 2026P2-DPO: Grounding Hallucination in Perceptual Processing via Calibration Direct Preference Optimization
  4. 2026Aligning with Your Own Voice: Self-Corrected Preference Learning for Hallucination Mitigation in LVLMs
  5. 2026Grounded Post-Training with Hard Examples for Reducing Hallucination in Multimodal Large Language Models
  6. 2025Mitigating Hallucinations in Large Vision-Language Models via Entity-Centric Multimodal Preference Optimization
  7. 2024Token Preference Optimization with Self-Calibrated Visual-Anchored Rewards for Hallucination Mitigation
  8. 2026DA-DPO: Cost-efficient Difficulty-aware Preference Optimization for Reducing MLLM Hallucinations