Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2602.14098 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914331319336960 |
|---|---|
| author | Wang, Youqi Chen, Shen Wang, Haowei Peng, Rongxuan Yao, Taiping Tan, Shunquan Chen, Changsheng Li, Bin Ding, Shouhong |
| author_facet | Wang, Youqi Chen, Shen Wang, Haowei Peng, Rongxuan Yao, Taiping Tan, Shunquan Chen, Changsheng Li, Bin Ding, Shouhong |
| contents | Existing Multimodal Large Language Models (MLLMs) for image forgery detection and localization predominantly operate under a text-centric Chain-of-Thought (CoT) paradigm. However, forcing these models to textually characterize imperceptible low-level tampering traces inevitably leads to hallucinations, as linguistic modalities are insufficient to capture such fine-grained pixel-level inconsistencies. To overcome this, we propose ForgeryVCR, a framework that incorporates a forensic toolbox to materialize imperceptible traces into explicit visual intermediates via Visual-Centric Reasoning. To enable efficient tool utilization, we introduce a Strategic Tool Learning post-training paradigm, encompassing gain-driven trajectory construction for Supervised Fine-Tuning (SFT) and subsequent Reinforcement Learning (RL) optimization guided by a tool utility reward. This paradigm empowers the MLLM to act as a proactive decision-maker, learning to spontaneously invoke multi-view reasoning paths including local zoom-in for fine-grained inspection and the analysis of invisible inconsistencies in compression history, noise residuals, and frequency domains. Extensive experiments reveal that ForgeryVCR achieves state-of-the-art (SOTA) performance in both detection and localization tasks, demonstrating superior generalization and robustness with minimal tool redundancy. The project page is available at https://youqiwong.github.io/projects/ForgeryVCR/. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_14098 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | ForgeryVCR: Visual-Centric Reasoning via Efficient Forensic Tools in MLLMs for Image Forgery Detection and Localization Wang, Youqi Chen, Shen Wang, Haowei Peng, Rongxuan Yao, Taiping Tan, Shunquan Chen, Changsheng Li, Bin Ding, Shouhong Computer Vision and Pattern Recognition Existing Multimodal Large Language Models (MLLMs) for image forgery detection and localization predominantly operate under a text-centric Chain-of-Thought (CoT) paradigm. However, forcing these models to textually characterize imperceptible low-level tampering traces inevitably leads to hallucinations, as linguistic modalities are insufficient to capture such fine-grained pixel-level inconsistencies. To overcome this, we propose ForgeryVCR, a framework that incorporates a forensic toolbox to materialize imperceptible traces into explicit visual intermediates via Visual-Centric Reasoning. To enable efficient tool utilization, we introduce a Strategic Tool Learning post-training paradigm, encompassing gain-driven trajectory construction for Supervised Fine-Tuning (SFT) and subsequent Reinforcement Learning (RL) optimization guided by a tool utility reward. This paradigm empowers the MLLM to act as a proactive decision-maker, learning to spontaneously invoke multi-view reasoning paths including local zoom-in for fine-grained inspection and the analysis of invisible inconsistencies in compression history, noise residuals, and frequency domains. Extensive experiments reveal that ForgeryVCR achieves state-of-the-art (SOTA) performance in both detection and localization tasks, demonstrating superior generalization and robustness with minimal tool redundancy. The project page is available at https://youqiwong.github.io/projects/ForgeryVCR/. |
| title | ForgeryVCR: Visual-Centric Reasoning via Efficient Forensic Tools in MLLMs for Image Forgery Detection and Localization |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2602.14098 |