# SafeScreen: A Safety-First Screening Framework for Personalized Video Retrieval for Vulnerable Users

Source: [SafeScreen: A Safety-First Screening Framework for Personalized Video Retrieval for Vulnerable Users](<https://www.sota2.com/research/paper/69d48f967d0172140ec74317>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-08T05:00:04.658Z

Authors: Wenzheng Zhao, Madhava Kalyan Gadiputi, Fengpei Yuan

Published: 2026-03-12T21:28:10.000Z

## Abstract

Open-domain video platforms offer rich, personalized content that could support health, caregiving, and educational applications, but their engagement-optimized recommendation algorithms can expose vulnerable users to inappropriate or harmful material. These risks are especially acute in child-directed and care settings (e.g., dementia care), where content must satisfy individualized safety constraints before being shown. We introduce SafeScreen, a safety-first video screening framework that retrieves and presents personalized video while enforcing individualized safety constraints. Rather than ranking videos by relevance or popularity, SafeScreen treats safety as a prerequisite and performs sequential approval or rejection of candidate videos through an automated pipeline. SafeScreen integrates three key components: (i) profile-driven extraction of individualized safety criteria, (ii) evidence-grounded assessments via adaptive question generation and multimodal VideoRAG analysis, and (iii) LLM-based decision-making that verifies safety, appropriateness, and relevance before content exposure. This design enables explainable, real-time screening of uncurated video repositories without relying on precomputed safety labels. We evaluate SafeScreen in a dementia-care reminiscence case study using 30 synthetic patient profiles and 90 test queries. Results demonstrate that SafeScreen prioritizes safety over engagement, diverging from YouTube's engagement-optimized rankings in 80-93% of cases, while maintaining high levels of safety coverage, sensibleness, and groundedness, as validated by both LLM-based evaluation and domain experts.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Video Safety Screening on SafeScreen Single-video cases (evaluation set)](<https://www.sota2.com/research/sota/video-safety-screening-on-safescreen-single-video-cases-evaluation-set.md>): Video safety screening; SafeScreen Single-video cases (evaluation set); YouTube Top Result Match Rate: 25.

- [Video Safety Screening on SafeScreen Multi-video cases](<https://www.sota2.com/research/sota/video-safety-screening-on-safescreen-multi-video-cases.md>): Video safety screening; SafeScreen Multi-video cases (evaluation); YouTube Match Rate (%): 14.3.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2604.03264>)

[PDF](<https://arxiv.org/pdf/2604.03264.pdf>)