OmniVL-Guard: Towards Unified Vision-Language Forgery Detection and Grounding via Balanced RL
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917498084917248 |
|---|---|
| author | Shen, Jinjie Wu, Jing Wang, Yaxiong Cheng, Lechao Tang, Shengeng Hui, Tianrui Pu, Nan Zhong, Zhun |
| author_facet | Shen, Jinjie Wu, Jing Wang, Yaxiong Cheng, Lechao Tang, Shengeng Hui, Tianrui Pu, Nan Zhong, Zhun |
| contents | Existing forgery detection methods are often limited to uni-modal or bi-modal settings, failing to handle the interleaved text, images, and videos prevalent in real-world misinformation. To bridge this gap, this paper targets to develop a unified framework for omnibus vision-language forgery detection and grounding. In this unified setting, the {interplay} between diverse modalities and the dual requirements of simultaneous detection and localization pose a critical ``difficulty bias`` problem: the simpler veracity classification task tends to dominate the gradients, leading to suboptimal performance in fine-grained grounding during multi-task optimization. To address this challenge, we propose \textbf{OmniVL-Guard}, a balanced reinforcement learning framework for omnibus vision-language forgery detection and grounding. Particularly, OmniVL-Guard comprises two core designs: Self-Evolving CoT Generatio and Adaptive Reward Scaling Policy Optimization (ARSPO). {Self-Evolving CoT Generation} synthesizes high-quality reasoning paths, effectively overcoming the cold-start challenge. Building upon this, {Adaptive Reward Scaling Policy Optimization (ARSPO)} dynamically modulates reward scales and task weights, ensuring a balanced joint optimization. Extensive experiments demonstrate that OmniVL-Guard significantly outperforms state-of-the-art methods and exhibits zero-shot robust generalization across out-of-domain scenarios. The dataset and code are publicly available at https://github.com/shen8424/OmniVL-Guard. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_10687 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | OmniVL-Guard: Towards Unified Vision-Language Forgery Detection and Grounding via Balanced RL Shen, Jinjie Wu, Jing Wang, Yaxiong Cheng, Lechao Tang, Shengeng Hui, Tianrui Pu, Nan Zhong, Zhun Computer Vision and Pattern Recognition Artificial Intelligence Existing forgery detection methods are often limited to uni-modal or bi-modal settings, failing to handle the interleaved text, images, and videos prevalent in real-world misinformation. To bridge this gap, this paper targets to develop a unified framework for omnibus vision-language forgery detection and grounding. In this unified setting, the {interplay} between diverse modalities and the dual requirements of simultaneous detection and localization pose a critical ``difficulty bias`` problem: the simpler veracity classification task tends to dominate the gradients, leading to suboptimal performance in fine-grained grounding during multi-task optimization. To address this challenge, we propose \textbf{OmniVL-Guard}, a balanced reinforcement learning framework for omnibus vision-language forgery detection and grounding. Particularly, OmniVL-Guard comprises two core designs: Self-Evolving CoT Generatio and Adaptive Reward Scaling Policy Optimization (ARSPO). {Self-Evolving CoT Generation} synthesizes high-quality reasoning paths, effectively overcoming the cold-start challenge. Building upon this, {Adaptive Reward Scaling Policy Optimization (ARSPO)} dynamically modulates reward scales and task weights, ensuring a balanced joint optimization. Extensive experiments demonstrate that OmniVL-Guard significantly outperforms state-of-the-art methods and exhibits zero-shot robust generalization across out-of-domain scenarios. The dataset and code are publicly available at https://github.com/shen8424/OmniVL-Guard. |
| title | OmniVL-Guard: Towards Unified Vision-Language Forgery Detection and Grounding via Balanced RL |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2602.10687 |