FACT:一种具有编译工具使用轨迹的取证代理,用于AI生成图像检测
FACT: A Forensic Agent with Compiled Tool-Use Trajectories for AI-Generated Image Detection
- Lingnan University(岭南大学)
- Harbin Institute of Technology, Shenzhen(哈尔滨工业大学(深圳))
- The University of Hong Kong(香港大学)
- The Chinese University of Hong Kong(香港中文大学)
- Shenzhen University(深圳大学)
机构由 AI 辅助整理,请以论文原文为准。
AI总结:
FACT提出一种基于工具使用轨迹的取证代理,通过演化-蒸馏-精炼流程学习图像条件策略,在多个基准上实现最优的AI生成图像检测性能。
AI中文摘要:
AI生成图像的检测日益呈现开放世界的特点:新的图像生成器产生高度逼真的图像,使得视觉伪影更难被识别。现有的检测器通常依赖一组固定的取证线索,因此对某一生成器家族有效的检测器可能在另一生成器上失效。我们引入了FACT(具有编译工具使用轨迹的取证代理),它学习一种以图像为条件的工具使用策略,用于取证分析。FACT不是应用固定的检测器,而是决定调用哪些取证工具,解释返回的证据,并在收集到足够证据时停止。FACT遵循“演化-蒸馏-精炼”流程:它演化出一个经过执行验证的取证技能,将该技能编译为动作-观察工具使用轨迹,将其蒸馏为一个紧凑的代理,并使用成本感知的GRPO精炼策略。在两个内部基准和四个公开基准上,FACT在所有比较方法中取得了最佳性能,包括在近期未见过的生成器、深度伪造和篡改图像上。
英文摘要:
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.