FACT: A Forensic Agent with Compiled Tool-Use Trajectories for AI-Generated Image Detection

arXiv:2609.05876v1 Announce Type: new Abstract: AI-generated image detection is increasingly open-world: new image generators produce highly realistic images that make visual artifacts harder to identify. Existing detectors usually rely on a fixed set of forensic cues, so a detector that works well for one generator family may fail on another. We introduce FACT (Forensic Agent with Compiled Tool-use Trajectories), which learns an image-conditioned tool-use policy for forensic analysis. Instead of applying a fixed detector, FACT decides which forensic tools to call, interprets the returned evidence, and stops when sufficient evidence has been collected. FACT follows an Evolve--Distill--Refine pipeline: it evolves an execution-verified forensic skill, compiles the skill into action--observation tool-use trajectories, distills them into a compact agent, and refines the policy with cost-aware GRPO. Across two internal and four public benchmarks, FACT achieves the best performance among all compared methods, including on recent unseen generators, deepfakes, and manipulated images.

Related

• ForensicConcept: Transferable Forensic Concepts for AIGI Detection
• OVOD-Agent: A Markov-Bandit Framework for Proactive Visual Reasoning and Self-Evolving Detection
• ReAlign: Generalizable Image Forgery Detection via Reasoning-Aligned Representation
• Veritas++: Value-aware On-Policy Distillation for Perception-Enhanced AIGI Detection

Source: arXiv cs.CV | 2026-09-09