Deception Archive: Standardized Verbal Deception Datasets

The Deception Archive is a curated repository of verbal deception datasets identified through a systematic review of the computational verbal deception literature. The Archive was developed through a structured workflow involving dataset identification, retrieval, manual inspection, metadata curation, and data standardization. The current release contains 42 standardized datasets organized according to a common data structure, while preserving the original textual content and veracity annotations and providing additional standardized metadata describing dataset characteristics. The Archive is designed to facilitate dataset discovery, exploration, cross-study comparison, and reuse in computational research on verbal deception.