ForeAgent combines a Perception-Verdict MLLM architecture with hindsight-driven self-refining via sampling-reflection-evolution to reach 82.18% accuracy on Chameleon and 93.3% mean accuracy across 16 generators on AIGCDetect-Benchmark.
Common sense reasoning for deep fake detection
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 4years
2026 4roles
background 1polarities
background 1representative citing papers
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.
MFVLR uses multi-domain vision-language reconstruction with a fine-grained language transformer, multi-domain vision encoder, and vision injection module to achieve generalizable detection and localization of diffusion-synthesized face forgeries.
citing papers explorer
-
Perception, Verdict, and Evolution: Hindsight-Driven Self-Refining Forensics Agent for AI-Generated Image Detection
ForeAgent combines a Perception-Verdict MLLM architecture with hindsight-driven self-refining via sampling-reflection-evolution to reach 82.18% accuracy on Chameleon and 93.3% mean accuracy across 16 generators on AIGCDetect-Benchmark.
-
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.
-
MFVLR: Multi-domain Fine-grained Vision-Language Reconstruction for Generalizable Diffusion Face Forgery Detection and Localization
MFVLR uses multi-domain vision-language reconstruction with a fine-grained language transformer, multi-domain vision encoder, and vision injection module to achieve generalizable detection and localization of diffusion-synthesized face forgeries.
- DVAR: Adversarial Multi-Agent Debate for Video Authenticity Detection