Saved in:
Bibliographic Details
Main Authors: Eo, Jeyeon, Kim, Joo Young, Ju, Ran, Jung, Minyoung, Lee, Unggi
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.28089
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916053592834048
author Eo, Jeyeon
Kim, Joo Young
Ju, Ran
Jung, Minyoung
Lee, Unggi
author_facet Eo, Jeyeon
Kim, Joo Young
Ju, Ran
Jung, Minyoung
Lee, Unggi
contents BuddyBench introduces a privacy-constrained multi-task benchmark for pediatric social-communication personalization. Unlike existing neurodevelopmental repositories that primarily emphasize imaging, genetics, or cross-sectional clinical phenotyping, BuddyBench links drill-level learning trajectories, standardized clinical assessments, BuddyPlan self-report, and randomized-treatment endpoints within a unified benchmark schema. BuddyBench combines two cohorts: ND-03 is an observational cohort with dense drill coverage for Tasks1-2 (n = 189), and ND-02 is a randomized controlled trial cohort for Tasks3-4 (n = 86 ITT). Together, they support knowledge tracing, next-drill recommendation, clinical prediction, and causal inference, linking behavioral personalization to clinical evaluation. We additionally introduce BuddyBench-Sim, a synthetic companion dataset for reproducible evaluation. Baselines show signal across tasks while keeping pediatric clinical records protected.
format Preprint
id arxiv_https___arxiv_org_abs_2605_28089
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle BuddyBench: A Privacy-Constrained Multi-Task Benchmark for Pediatric Social-Communication Personalization
Eo, Jeyeon
Kim, Joo Young
Ju, Ran
Jung, Minyoung
Lee, Unggi
Artificial Intelligence
BuddyBench introduces a privacy-constrained multi-task benchmark for pediatric social-communication personalization. Unlike existing neurodevelopmental repositories that primarily emphasize imaging, genetics, or cross-sectional clinical phenotyping, BuddyBench links drill-level learning trajectories, standardized clinical assessments, BuddyPlan self-report, and randomized-treatment endpoints within a unified benchmark schema. BuddyBench combines two cohorts: ND-03 is an observational cohort with dense drill coverage for Tasks1-2 (n = 189), and ND-02 is a randomized controlled trial cohort for Tasks3-4 (n = 86 ITT). Together, they support knowledge tracing, next-drill recommendation, clinical prediction, and causal inference, linking behavioral personalization to clinical evaluation. We additionally introduce BuddyBench-Sim, a synthetic companion dataset for reproducible evaluation. Baselines show signal across tasks while keeping pediatric clinical records protected.
title BuddyBench: A Privacy-Constrained Multi-Task Benchmark for Pediatric Social-Communication Personalization
topic Artificial Intelligence
url https://arxiv.org/abs/2605.28089