# Can Vision-Language Models Reason about AI Edits in Images?

arXiv: https://arxiv.org/abs/2607.28464 · HTML: https://mlsanity.com/paper/2607.28464/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 10 of its 34 references that are in the map.
- First posted on arXiv 30 Jul 2026.

## Placed in

- LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/)
- Multimodal LLMs / vision-language models / [Embodied and 3D vision-language models](https://mlsanity.com/mllm/embodied-and-3d-vision-language-models/)

## Builds on (most cited first, 10 of 10)

- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Gemini: A Family of Highly Capable Multimodal Models](https://mlsanity.com/paper/2312.11805/) (2023)
- [A Survey on Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2402.00253/) (2024)
- [Ivy-Fake: A Unified Explainable Framework and Benchmark for Image and Video AIGC Detection](https://mlsanity.com/paper/2506.00979/) (2025)
- [ForgeryGPT: A Multimodal LLM for Interpretable Image Forgery Detection and Localization](https://mlsanity.com/paper/2410.10238/) (2024)
- [ForgerySleuth: Empowering Multimodal Large Language Models for Image Manipulation Detection](https://mlsanity.com/paper/2411.19466/) (2024)
- [Fake-HR1: Rethinking Reasoning of Vision Language Model for Synthetic Image Detection](https://mlsanity.com/paper/2602.10042/) (2026)
- [ThinkFake: Reasoning in Multimodal Large Language Models for AI-Generated Image Detection](https://mlsanity.com/paper/2509.19841/) (2025)
- [Comprehensive Dataset of Face Manipulations for Development and Evaluation of Forensic Tools](https://mlsanity.com/paper/2208.11776/) (2022)

## Abstract (arXiv)

Detection and localization of AI-tampered images are critical for trustworthy AI, yet modern generative models have made such manipulations increasingly difficult to identify. While traditional binary classifiers can detect image tampering, they lack interpretability and generalization. Vision-Language Models (VLMs) offer a promising alternative due to their strong visual understanding and reasoning capabilities; however, existing approaches typically rely on supervised finetuning with curated explanations rather than exploiting their inherent reasoning capabilities. In this work, we investigate whether VLMs can be trained to reason about AI-generated image edits using reinforcement learning (RL) rather than explicit reasoning supervision. Motivated by the success in Group Relative Policy Optimization (GRPO), an RL technique that incentivizes the model to reason by generating thinking traces prior to giving the final answer, we propose a GRPO-based training framework that utilizes simple accuracy and format rewards. Given an input image, the model produces a structured reasoning trace and predicts whether the image has been tampered with. A lightweight segmentation model is then guided by the reasoning output to generate pixel-level localization masks. Experiments across multiple image manipulation datasets demonstrate that our approach achieves competitive detection and localization performance compared to state-of-the-art image forgery detectors, despite requiring substantially weaker supervision. We introduce effective intersection over union (eff-IoU), a unified metric to jointly evaluate detection and localization. These results suggest that reinforcement learning provides an effective and scalable mechanism for teaching VLMs to reason about AI-generated content.