Open-World / Open-Vocabulary From survey PDF — verify links
PLOVAD
PLOVAD — Prompt Adaptation
IEEE TCSVT · 2025Human-reviewed summary status
Survey summary
PLOVAD prompt-tunes frozen image-based vision-language models for open-vocabulary VAD via a Prompting Module -- a learnable domain-specific prompt plus an LLM-crafted anomaly-specific prompt -- and a Temporal Module that stacks a graph attention network atop frame-wise visual features for video-level localization.
Taxonomy placement
Open-World
Open-WorldOpen-VocabularyPrompt Adaptation
Method characteristics
Open-World / Open-VocabularyDomain promptLLM anomaly promptGAT temporal