A dual-encoder deepfake detector pairs a frozen specialist with a LoRA-tuned MLLM, trained first via binary alignment then via RL to reward explain-then-classify behavior, yielding improved cross-dataset performance and interpretability.
ArXivabs/2406.08625(2024),https: //api.semanticscholar.org/CorpusID:2704405863
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
S^3 extracts dominant shortcut directions from a linear forgery-method classifier using SVD and attenuates them in feature space to improve cross-method generalization in deepfake detection.
citing papers explorer
-
The Regularizing Power of Language-Training Deepfake Detectors
A dual-encoder deepfake detector pairs a frozen specialist with a LoRA-tuned MLLM, trained first via binary alignment then via RL to reward explain-then-classify behavior, yielding improved cross-dataset performance and interpretability.
-
Suppressing Forgery-Specific Shortcuts for Generalizable Deepfake Detection
S^3 extracts dominant shortcut directions from a linear forgery-method classifier using SVD and attenuates them in feature space to improve cross-method generalization in deepfake detection.