mDPO: Conditional Preference Optimization for Multimodal Large Language Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909338069630976 |
|---|---|
| author | Wang, Fei Zhou, Wenxuan Huang, James Y. Xu, Nan Zhang, Sheng Poon, Hoifung Chen, Muhao |
| author_facet | Wang, Fei Zhou, Wenxuan Huang, James Y. Xu, Nan Zhang, Sheng Poon, Hoifung Chen, Muhao |
| contents | Direct preference optimization (DPO) has shown to be an effective method for large language model (LLM) alignment. Recent works have attempted to apply DPO to multimodal scenarios but have found it challenging to achieve consistent improvement. Through a comparative experiment, we identify the unconditional preference problem in multimodal preference optimization, where the model overlooks the image condition. To address this problem, we propose mDPO, a multimodal DPO objective that prevents the over-prioritization of language-only preferences by also optimizing image preference. Moreover, we introduce a reward anchor that forces the reward to be positive for chosen responses, thereby avoiding the decrease in their likelihood -- an intrinsic problem of relative preference optimization. Experiments on two multimodal LLMs of different sizes and three widely used benchmarks demonstrate that mDPO effectively addresses the unconditional preference problem in multimodal preference optimization and significantly improves model performance, particularly in reducing hallucination. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_11839 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | mDPO: Conditional Preference Optimization for Multimodal Large Language Models Wang, Fei Zhou, Wenxuan Huang, James Y. Xu, Nan Zhang, Sheng Poon, Hoifung Chen, Muhao Computer Vision and Pattern Recognition Artificial Intelligence Computation and Language Machine Learning Direct preference optimization (DPO) has shown to be an effective method for large language model (LLM) alignment. Recent works have attempted to apply DPO to multimodal scenarios but have found it challenging to achieve consistent improvement. Through a comparative experiment, we identify the unconditional preference problem in multimodal preference optimization, where the model overlooks the image condition. To address this problem, we propose mDPO, a multimodal DPO objective that prevents the over-prioritization of language-only preferences by also optimizing image preference. Moreover, we introduce a reward anchor that forces the reward to be positive for chosen responses, thereby avoiding the decrease in their likelihood -- an intrinsic problem of relative preference optimization. Experiments on two multimodal LLMs of different sizes and three widely used benchmarks demonstrate that mDPO effectively addresses the unconditional preference problem in multimodal preference optimization and significantly improves model performance, particularly in reducing hallucination. |
| title | mDPO: Conditional Preference Optimization for Multimodal Large Language Models |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2406.11839 |