ML Security PapersAIFIND: Artifact-Aware Interpreting Fine-Grained Alignment for Incremental Face Forgery Detection
Hao Wang , Beichen Zhang , Yanpei Gong , Shaoyi Fang , Zhaobo Qi , Yuanrong Xu , Xinyan Liu , Weigang Zhang
Published on arXiv
2604.16207
Output Integrity Attack
OWASP ML Top 10 — ML09
Key Finding
Achieves superior performance on multiple incremental learning protocols by stabilizing feature space through semantic anchor alignment
AIFIND
Novel technique introduced
As forgery types continue to emerge consistently, Incremental Face Forgery Detection (IFFD) has become a crucial paradigm. However, existing methods typically rely on data replay or coarse binary supervision, which fails to explicitly constrain the feature space, leading to severe feature drift and catastrophic forgetting. To address this, we propose AIFIND, Artifact-Aware Interpreting Fine-Grained Alignment for Incremental Face Forgery Detection, which leverages semantic anchors to stabilize incremental learning. We design the Artifact-Driven Semantic Prior Generator to instantiate invariant semantic anchors, establishing a fixed coordinate system from low-level artifact cues. These anchors are injected into the image encoder via Artifact-Probe Attention, which explicitly constrains volatile visual features to align with stable semantic anchors. Adaptive Decision Harmonizer harmonizes the classifiers by preserving angular relationships of semantic anchors, maintaining geometric consistency across tasks. Extensive experiments on multiple incremental protocols validate the superiority of AIFIND.
Key Contributions
- Artifact-Driven Semantic Prior Generator that creates invariant semantic anchors from low-level forgery artifacts
- Artifact-Probe Attention mechanism that aligns visual features with stable semantic anchors to prevent feature drift
- Data-replay-free incremental learning paradigm for face forgery detection that mitigates catastrophic forgetting
🛡️ Threat Analysis
Primary contribution is detecting AI-generated/manipulated face images (deepfakes). The paper addresses face forgery detection — authenticating whether facial images are real or forged. This is output integrity and content authenticity verification, core to ML09.