A five-stage pipeline is proposed to detect and explain AI-generated 32x32 images by localizing and describing visual artifacts, with only qualitative examples as evidence.
VisualGPT: Data-efficient Adaptation of Pre- trained Language Models for Image Captioning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
VERITAS: Verification and Explanation of Realness in Images for Transparency in AI Systems
A five-stage pipeline is proposed to detect and explain AI-generated 32x32 images by localizing and describing visual artifacts, with only qualitative examples as evidence.