MM-VeriAgent: Learning to Use Extensive Tools to Verify Multimodal Misinformation with Reinforcement Learning
cs.CV
Submitted: 2026-09-25
Updated: 2026-09-25
License: http://creativecommons.org/licenses/by/4.0/
The gist: Real-world multimodal misinformation often involves mixed forgery sources, requiring sample-specific detection strategies.
Terminology
Abstract
Real-world multimodal misinformation often involves mixed forgery sources, requiring sample-specific detection strategies. Existing tool-augmented methods rely on predefined workflows or inference-time planning, limiting adaptability or increasing inference cost. To address this issue, we introduce MM-VeriAgent, which learns to verify mixed-source multimodal misinformation with tools. We first build MM-VeriTools, a specialized toolkit for misinformation detection agents. By benchmarking various candidate models and methods on the sub-tasks required by mixed-source detection, we select the strongest for textual, visual, and cross-modal forgery analysis and encapsulate them as callable tools with a unified interface. On top of this toolkit, we train the LVLM agent with reinforcement learning to teach it how to use these tools to better solve mixed-source detection. Since many of the tools are specialized models whose online execution at every rollout severely limits RL efficiency, we further introduce Tool-Execution Cache, which pre-executes candidate tool calls and reuses their cached outputs during training. This preserves multi-step rollouts while reducing online tool execution, largely improving the training efficiency.Experiments on MMFakeBench demonstrate substantial accuracy gains over the base model without explicit tool search at inference time. Ablation and efficiency analyses further validate the learned tool-use policy and show that Tool-Execution Cache reduces online tool executions during training.
Sources
- Qwen2.5-VL Technical Report
- Qwen3-VL Technical Report
- Can LLMs Improve Multimodal Fact-Checking by Asking Relevant Questions?
- Dual Data Alignment Makes AI-Generated Image Detector Easier Generalizable
- Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
- Tool-Star: Empowering LLM-Brained Multi-Tool Reasoner via Reinforcement Learning
- ReTool: Reinforcement Learning for Strategic Tool Use in LLMs
- Veri-R1: Toward Precise and Faithful Claim Verification via Online Reinforcement Learning
- CacheRL:Multi-Turn Tool-Calling Agents via Cached Rollouts and Hybrid Reward
- VerlTool: Towards Holistic Agentic Reinforcement Learning with Tool Use
- Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning
- Large Language Model Agent for Fake News Detection
- Simulating Environments with Reasoning Models for Agent Training
- GPT-4o System Card
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- TVCACHE: A Stateful Tool-Value Cache for Post-Training LLM Agents
- Qwen3 Technical Report
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models