AVI Challenge 2026: Assessing True Personality Traits and Cognitive Ability from Asynchronous Video Interviews (AVIs)

Tianyi Zhang, Tianhua Qi, Reinout E. de Vries, Wenming ZHENG, Janneke Oostrom, Djurre Holtrop, Yuan Zong, Antonios Koutsoumpis

Keywords: Personality Assessment, Interview Performance, Cognitive Ability Assessment, Asynchronous Video Interviews

Abstract: Asynchronous Video Interviews (AVIs) allow candidates to record responses to predefined questions using digital devices, offering flexibility and enabling remote assessments. Evaluating personality traits through AVIs provides organizations with valuable insights into candidates’ profiles and helps predicting future job performance. Building on this, we successfully organized the AVI Challenge 2025 at ACM Multimedia 2025, which established a high-fidelity benchmark by grounding assessments in Trait Activation Theory and employing Behaviorally Anchored Rating Scales (BARS). However, the AVI Challenge 2025 primarily focused on observer-reported personality, which captures "ascribed" traits susceptible to human perceptual bias and candidates' impression management. In addition, it bypass cognitive ability, a key predictor of job performance. To address these gaps, we introduce the "AVI Challenge 2026", featuring a novel dataset of mock AVIs (3,876 videos from 646 subjects). This iteration introduces two major shifts: (a) transitioning from social perception to "true" personality by utilizing validated self-reported inventories as ground truth, and (b) inaugurating a new track for cognitive ability estimation through semantic and paralinguistic analysis. By aligning multimedia analytics with the "gold standards" of psychology, this challenge aims to decode authentic professional potential. Additionally, ongoing data collection from Dutch and Mandarin speakers will make the dataset cross-cultural and multilingual, broadening its utility for global recruitment applications.

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 2