EUVIMEDEUVIMED entdecken
Crossref

Open-World Deepfake Attribution via Confidence-Aware Asymmetric Learning

Haiyang Zheng, Nan Pu, Wenjing Li, Teng Long, Nicu Sebe, Zhun Zhong

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (16) · 13378-13386

DOI: 10.1609/aaai.v40i16.38341 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Haiyang Zheng und 5 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The proliferation of synthetic facial imagery has intensified the need for robust Open-World DeepFake Attribution (OW-DFA), which aims to attribute both known and unknown forgeries using labeled data for known types and unlabeled data containing a mixture of known and novel types. However, existing OW-DFA methods face two critical limitations: 1) A confidence skew that leads to unreliable pseudo-labels for novel forgeries, resulting in biased training. 2) An unrealistic assumption that the number of unknown forgery types is known a priori. To address these challenges, we propose a Confidence-aware Asymmetric Learning (CAL) framework, which adaptively balances model confidence across known and novel forgery types. CAL mainly consists of two components: Confidence-aware Consistency Regularization (CCR) and Asymmetric Confidence Reinforcement (ACR). CCR mitigates pseudo-label bias by dynamically scaling sample losses based on normalized confidence, gradually shifting the training focus from high- to low-confidence samples. ACR complements this by separately calibrating confidence for known and novel classes through selective learning on high-confidence samples, guided by their confidence gap. Together, CCR and ACR form a mutually reinforcing loop that significantly improves the model's OW-DFA performance. Moreover, we introduce a Dynamic Prototype Pruning (DPP) strategy that automatically estimates the number of novel forgery types in a coarse-to-fine manner, removing the need for unrealistic prior assumptions and enhancing the scalability of our methods to real-world OW-DFA scenarios. Extensive experiments on the standard and OW-DFA benchmark and a newly extended benchmark incorporating advanced manipulations demonstrate that CAL consistently outperforms previous methods, achieving new state-of-the-art performance on both known and novel forgery attribution.

Zitieren

APA

Zheng, H., Pu, N., Li, W., Long, T., Sebe, N., & Zhong, Z. (2026). Open-World Deepfake Attribution via Confidence-Aware Asymmetric Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 40(16), 13378-13386. https://doi.org/10.1609/aaai.v40i16.38341

Vancouver

Zheng H, Pu N, Li W, Long T, Sebe N, Zhong Z. Open-World Deepfake Attribution via Confidence-Aware Asymmetric Learning. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(16):13378-13386. doi:10.1609/aaai.v40i16.38341

BibTeX
@article{Zheng2026,
  author = {Zheng, Haiyang and Pu, Nan and Li, Wenjing and Long, Teng and Sebe, Nicu and Zhong, Zhun},
  title = {Open-World Deepfake Attribution via Confidence-Aware Asymmetric Learning},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {16},
  pages = {13378--13386},
  doi = {10.1609/aaai.v40i16.38341}
}
Kostenlos starten →