FACT: A Forensic Agent with Compiled Tool-Use Trajectories for AI-Generated Image Detection
Authors: Jiaoyang Chen, Bin Hu, Jingyu Hu, Kun Zhou, Qin Zhang, Zhengzhe Liu
Organizations: Lingnan University, Hong Kong · Harbin Institute of Technology, Shenzhen · The University of Hong Kong · The Chinese University of Hong Kong · Shenzhen University
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.
Figures & tables
Figure 1: Can you tell which images are real and which are generated by AI? 2 2 2 In Fig. 2 , the first two images are AI-generated, while the last two are real photographs.
Method
Nano Banana
Nano Pro
GPT- Image-1
SD3.5 L
SD3 M
Z-Image Turbo
VERITAS
0.74
0.75
0.76
0.50
0.64
0.35
SIDA
0.70
0.61
0.51
0.53
0.58
0.57
SDAIE
0.64
0.54
0.36
0.69
0.70
1.00
AIDE
0.52
0.84
0.55
0.49
0.50
0.98
Comm.
0.67
0.56
0.59
0.52
0.83
0.71
Table 1: Detector specialization across recent generators. Each column is a balanced real-versus-synthetic subset. “Comm.” denotes Community Forensics; L and M denote Large and Medium. Darker cells indicate higher accuracy, and bold values mark the best detector in each subset.
Figure 2: Overview of FACT. FACT trains a forensic agent with an Evolve–Distill–Refine pipeline. Skill evolution learns an executable investigation procedure from judged cases; trajectory distillation converts the skill into supervised action–observation tool-use data; and cost-aware GRPO teaches the agent when additional evidence is worth its cost. At inference, the trained agent adaptively calls forensic tools or stops with a final real/AI decision.
Method
STD3K
NewGen-900
Community ( Park and Owens 2024 )
78.56
72.67
Veritas ( Tan et al. 2025 )
40.78
43.11
SIDA ( Huang et al. 2024 )
77.38
61.00
SDAIE ( Zhong et al. 2025 )
59.53
76.89
AIDE ( Yan et al. 2025a )
44.31
73.44
Majority Vote
72.97
75.33
Table 2: Comparison on the internal benchmarks. All values are accuracies in percentages. All methods are evaluated using our fixed evaluation pipeline.
Method
Accuracy
NPR † ( Tan et al. 2024 )
57.81
AIDE † ( Yan et al. 2025a )
65.77
Community ( Park and Owens 2024 )
78.44
Veritas ( Tan et al. 2025 )
58.23
SIDA ( Huang et al. 2024 )
66.50
SDAIE ( Zhong et al. 2025 )
58.86
Table 3: Comparison on Chameleon. † : Table 5 of ( Yan et al. 2025a ) ; ‡ : cited method papers. Unmarked results use our fixed pipeline.
Method
BAcc.
Effort † ( Yan et al. 2025b )
74.09
FakeVLM † ( Wen et al. 2025 )
81.62
MIRROR † ( Liu et al. 2026 )
85.23
AIDE † ( Yan et al. 2025a )
73.70
FakeShield † ( Xu et al. 2025 )
61.55
SIDA † ( Huang et al. 2024 )
61.92
Table 4: Comparison on LOKI. † : Table 2 of ( Zhu et al. 2026 ) . Unmarked results use our fixed pipeline.
Method
Accuracy
AIDE † ( Yan et al. 2025a )
92.80
FakeVLM † ( Wen et al. 2025 )
81.00
UniShield † ( Huang et al. 2025 )
94.20
Community ( Park and Owens 2024 )
96.80
Veritas ( Tan et al. 2025 )
69.60
SIDA ( Huang et al. 2024 )
59.90
Table 5: Comparison on AIGCDetectionBenchmark. † : Table 6 of ( Huang et al. 2025 ) ; ‡ : cited method paper. Unmarked results use our fixed pipeline.
Table 6: Comparison on HydraFake. Values are average accuracies in percentages. † : Table 1 of ( Tan et al. 2025 ) ; unmarked results use our fixed pipeline. “T” abbreviates Think.
(a) Training stages Var. S D R STD New Direct – – – 87.90 50.80 Skill ✓ – – 89.40 75.10 Distill. ✓ ✓ – 91.53 81.89 Refine ✓ ✓ ✓ 91.40 82.44
Table 7: Pipeline and toolbox ablations. In (a), S, D, and R denote skill learning, trajectory distillation, and GRPO refinement; the variants are Direct, Skill-guided, Distilled, and Refined. In (b), Sig. and Exp. denote signal probes and model-based experts.
Task
Fmt.
Valid
Cost
Acc.
Acq.
✓
–
–
–
81.78
1.5300
✓
✓
–
–
82.00
1.5700
✓
✓
✓
–
82.56
1.5500
✓
✓
✓
✓
82.44
1.3835
Table 8: Reward-component ablation on NewGen-900. Reward terms are added cumulatively; the full reward uses λ=0.01 .
AI-generated image detection identifies synthetic or AI-edited images and develops methods that generalize across generators, domains, and image transformations.