English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deception detection in conversations is a challenging yet important task,having pivotal applications in many fields such as credibility assessment inbusiness, multimedia anti-frauds, and custom security. Despite this, deceptiondetection research is hindered by the lack of high-quality deception datasets,as well as the difficulties of learning multimodal features effectively. Toaddress this issue, we introduce DOLOS\footnote {The name ``DOLOS" comes fromGreek mythology.}, the largest gameshow deception detection dataset with richdeceptive conversations. DOLOS includes 1,675 video clips featuring 213subjects, and it has been labeled with audio-visual feature annotations. Weprovide train-test, duration, and gender protocols to investigate the impact ofdifferent factors. We benchmark our dataset on previously proposed deceptiondetection approaches. To further improve the performance by fine-tuning fewerparameters, we propose Parameter-Efficient Crossmodal Learning (PECL), where aUniform Temporal Adapter (UT-Adapter) explores temporal attention intransformer-based architectures, and a crossmodal fusion module, Plug-inAudio-Visual Fusion (PAVF), combines crossmodal information from audio-visualfeatures. Based on the rich fine-grained audio-visual annotations on DOLOS, wealso exploit multi-task learning to enhance performance by concurrentlypredicting deception and audio-visual features. Experimental resultsdemonstrate the desired quality of the DOLOS dataset and the effectiveness ofthe PECL. The DOLOS dataset and the source codes are available athttps://github.com/NMS05/Audio-Visual-Deception-Detection-DOLOS-Dataset-and-Parameter-Efficient-Crossmodal-Learning/tree/main.

Audio-Visual Deception Detection: DOLOS Dataset and Parameter-Efficient Crossmodal Learning