# Cultivating Forensic Reasoning for Generalizable Multimodal Manipulation Detection

arXiv: https://arxiv.org/abs/2603.01993 · HTML: https://mlsanity.com/paper/2603.01993/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 38 references that are in the map.
- First posted on arXiv 2 Mar 2026.

## Builds on (most cited first, 23 of 23)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://mlsanity.com/paper/2305.06500/) (2023)
- [FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning](https://mlsanity.com/paper/2307.08691/) (2023)
- [InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency](https://mlsanity.com/paper/2508.18265/) (2025)
- [ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision](https://mlsanity.com/paper/2102.03334/) (2021)
- [TinyBERT: Distilling BERT for Natural Language Understanding](https://mlsanity.com/paper/1909.10351/) (2019)
- [SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training](https://mlsanity.com/paper/2501.17161/) (2025)
- [DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding](https://mlsanity.com/paper/2412.10302/) (2024)
- [PandaGPT: One Model To Instruction-Follow Them All](https://mlsanity.com/paper/2305.16355/) (2023)
- [Detecting and Grounding Multi-Modal Media Manipulation](https://mlsanity.com/paper/2304.02556/) (2023)
- [Self-Rewarding Vision-Language Model via Reasoning Decomposition](https://mlsanity.com/paper/2508.19652/) (2025)
- [MMFakeBench: A Mixed-Source Multimodal Misinformation Detection Benchmark for LVLMs](https://mlsanity.com/paper/2406.08772/) (2024)
- [Detecting and Grounding Multi-Modal Media Manipulation and Beyond](https://mlsanity.com/paper/2309.14203/) (2023)
- [FKA-Owl: Advancing Multimodal Fake News Detection through Knowledge-Augmented LVLMs](https://mlsanity.com/paper/2403.01988/) (2024)
- [ASAP: Advancing Semantic Alignment Promotes Multi-Modal Manipulation Detecting and Grounding](https://mlsanity.com/paper/2412.12718/) (2024)
- [Challenges and Opportunities in Information Manipulation Detection: An Examination of Wartime Russian Media](https://mlsanity.com/paper/2205.12382/) (2022)
- [TRUST-VL: An Explainable News Assistant for General Multimodal Misinformation Detection](https://mlsanity.com/paper/2509.04448/) (2025)
- [The Coherence Trap: When MLLM-Crafted Narratives Exploit Manipulated Visual Contexts](https://mlsanity.com/paper/2505.17476/) (2025)
- [FakeSV-VLM: Taming VLM for Detecting Fake Short-Video News via Progressive Mixture-Of-Experts Adapter](https://mlsanity.com/paper/2508.19639/) (2025)
- [Generating Attribution Reports for Manipulated Facial Images: A Dataset and Baseline](https://mlsanity.com/paper/2412.19685/) (2024)
- [Beyond Artificial Misalignment: Detecting and Grounding Semantic-Coordinated Multimodal Manipulations](https://mlsanity.com/paper/2509.12653/) (2025)

## Built on by (most cited first, 2 of 2)

- [GeoMag: Geometric-Aware Video Motion Magnification via State Space Model](https://mlsanity.com/paper/2605.29762/) (2026)
- [Specialist-Generalist Fusion with Outcome-Supervised Rationales for Deepfake Detection](https://mlsanity.com/paper/2605.31192/) (2026)

## Abstract (arXiv)

Recent advances in generative AI have significantly enhanced the realism of multimodal media manipulation, thereby posing substantial challenges to manipulation detection. Existing manipulation detection and grounding approaches predominantly focus on manipulation type classification under result-oriented supervision, which not only lacks interpretability but also tends to overfit superficial artifacts. In this paper, we argue that generalizable detection requires incorporating explicit forensic reasoning, rather than merely classifying a limited set of manipulation types, which fails to generalize to unseen manipulation patterns. To this end, we propose REFORM, a reasoning-driven framework that shifts learning from outcome fitting to process modeling. REFORM adopts a three-stage curriculum that first induces forensic rationales, then aligns reasoning with final judgments, and finally refines logical consistency via reinforcement learning. To support this paradigm, we introduce ROM, a large-scale dataset with rich reasoning annotations. Extensive experiments show that REFORM establishes new state-of-the-art performance with superior generalization, achieving 81.52% ACC on ROM, 76.65% ACC on DGM4, and 74.9 F1 on MMFakeBench.