Unlocking the Essence of Beauty: Advanced Aesthetic Reasoning with Relative-Absolute Policy Optimization
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909808509059072 |
|---|---|
| author | Liu, Boyang Hu, Yifan Jin, Senjie Dou, Shihan Shi, Gonglei Shao, Jie Gui, Tao Huang, Xuanjing |
| author_facet | Liu, Boyang Hu, Yifan Jin, Senjie Dou, Shihan Shi, Gonglei Shao, Jie Gui, Tao Huang, Xuanjing |
| contents | Multimodal large language models (MLLMs) are well suited to image aesthetic assessment, as they can capture high-level aesthetic features leveraging their cross-modal understanding capacity. However, the scarcity of multimodal aesthetic reasoning data and the inherently subjective nature of aesthetic judgment make it difficult for MLLMs to generate accurate aesthetic judgments with interpretable rationales. To this end, we propose Aes-R1, a comprehensive aesthetic reasoning framework with reinforcement learning (RL). Concretely, Aes-R1 integrates a pipeline, AesCoT, to construct and filter high-quality chain-of-thought aesthetic reasoning data used for cold-start. After teaching the model to generate structured explanations prior to scoring, we then employ the Relative-Absolute Policy Optimization (RAPO), a novel RL algorithm that jointly optimizes absolute score regression and relative ranking order, improving both per-image accuracy and cross-image preference judgments. Aes-R1 enables MLLMs to generate grounded explanations alongside faithful scores, thereby enhancing aesthetic scoring and reasoning in a unified framework. Extensive experiments demonstrate that Aes-R1 improves the backbone's average PLCC/SRCC by 47.9%/34.8%, surpassing state-of-the-art baselines of similar size. More ablation studies validate Aes-R1's robust generalization under limited supervision and in out-of-distribution scenarios. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_21871 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Unlocking the Essence of Beauty: Advanced Aesthetic Reasoning with Relative-Absolute Policy Optimization Liu, Boyang Hu, Yifan Jin, Senjie Dou, Shihan Shi, Gonglei Shao, Jie Gui, Tao Huang, Xuanjing Computer Vision and Pattern Recognition Artificial Intelligence Multimodal large language models (MLLMs) are well suited to image aesthetic assessment, as they can capture high-level aesthetic features leveraging their cross-modal understanding capacity. However, the scarcity of multimodal aesthetic reasoning data and the inherently subjective nature of aesthetic judgment make it difficult for MLLMs to generate accurate aesthetic judgments with interpretable rationales. To this end, we propose Aes-R1, a comprehensive aesthetic reasoning framework with reinforcement learning (RL). Concretely, Aes-R1 integrates a pipeline, AesCoT, to construct and filter high-quality chain-of-thought aesthetic reasoning data used for cold-start. After teaching the model to generate structured explanations prior to scoring, we then employ the Relative-Absolute Policy Optimization (RAPO), a novel RL algorithm that jointly optimizes absolute score regression and relative ranking order, improving both per-image accuracy and cross-image preference judgments. Aes-R1 enables MLLMs to generate grounded explanations alongside faithful scores, thereby enhancing aesthetic scoring and reasoning in a unified framework. Extensive experiments demonstrate that Aes-R1 improves the backbone's average PLCC/SRCC by 47.9%/34.8%, surpassing state-of-the-art baselines of similar size. More ablation studies validate Aes-R1's robust generalization under limited supervision and in out-of-distribution scenarios. |
| title | Unlocking the Essence of Beauty: Advanced Aesthetic Reasoning with Relative-Absolute Policy Optimization |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2509.21871 |