Saved in:
Bibliographic Details
Main Authors: Panchal, Viraj, Talsaniya, Tanmay, Patel, Parag, Patel, Meet
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.16181
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914401870675968
author Panchal, Viraj
Talsaniya, Tanmay
Patel, Parag
Patel, Meet
author_facet Panchal, Viraj
Talsaniya, Tanmay
Patel, Parag
Patel, Meet
contents We present KidsNanny, a two-stage multimodal content moderation architecture for child safety. Stage 1 combines a vision transformer (ViT) with an object detector for visual screening (11.7 ms); outputs are routed as text not raw pixels to Stage 2, which applies OCR and a text based 7B language model for contextual reasoning (120 ms total pipeline). We evaluate on the UnsafeBench Sexual category (1,054 images) under two regimes: vision-only, isolating Stage 1, and multimodal, evaluating the full Stage 1+2 pipeline. Stage 1 achieves 80.27% accuracy and 85.39% F1 at 11.7 ms; vision-only baselines range from 59.01% to 77.04% accuracy. The full pipeline achieves 81.40% accuracy and 86.16% F1 at 120 ms, compared to ShieldGemma-2 (64.80% accuracy, 1,136 ms) and LlavaGuard (80.36% accuracy, 4,138 ms). To evaluate text-awareness, we filter two subsets: a text+visual subset (257 images) and a text-only subset (44 images where safety depends primarily on embedded text). On text-only images, KidsNanny achieves 100% recall (25/25 positives; small sample) and 75.76% precision; ShieldGemma-2 achieves 84% recall and 60% precision at 1,136 ms. Results suggest that dedicated OCR-based reasoning may offer recall-precision advantages on text-embedded threats at lower latency, though the small text-only subset limits generalizability. By documenting this architecture and evaluation methodology, we aim to contribute to the broader research effort on efficient multimodal content moderation for child safety.
format Preprint
id arxiv_https___arxiv_org_abs_2603_16181
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle KidsNanny: A Two-Stage Multimodal Content Moderation Pipeline Integrating Visual Classification, Object Detection, OCR, and Contextual Reasoning for Child Safety
Panchal, Viraj
Talsaniya, Tanmay
Patel, Parag
Patel, Meet
Computer Vision and Pattern Recognition
Cryptography and Security
I.4.9; I.2.7; K.4.1
We present KidsNanny, a two-stage multimodal content moderation architecture for child safety. Stage 1 combines a vision transformer (ViT) with an object detector for visual screening (11.7 ms); outputs are routed as text not raw pixels to Stage 2, which applies OCR and a text based 7B language model for contextual reasoning (120 ms total pipeline). We evaluate on the UnsafeBench Sexual category (1,054 images) under two regimes: vision-only, isolating Stage 1, and multimodal, evaluating the full Stage 1+2 pipeline. Stage 1 achieves 80.27% accuracy and 85.39% F1 at 11.7 ms; vision-only baselines range from 59.01% to 77.04% accuracy. The full pipeline achieves 81.40% accuracy and 86.16% F1 at 120 ms, compared to ShieldGemma-2 (64.80% accuracy, 1,136 ms) and LlavaGuard (80.36% accuracy, 4,138 ms). To evaluate text-awareness, we filter two subsets: a text+visual subset (257 images) and a text-only subset (44 images where safety depends primarily on embedded text). On text-only images, KidsNanny achieves 100% recall (25/25 positives; small sample) and 75.76% precision; ShieldGemma-2 achieves 84% recall and 60% precision at 1,136 ms. Results suggest that dedicated OCR-based reasoning may offer recall-precision advantages on text-embedded threats at lower latency, though the small text-only subset limits generalizability. By documenting this architecture and evaluation methodology, we aim to contribute to the broader research effort on efficient multimodal content moderation for child safety.
title KidsNanny: A Two-Stage Multimodal Content Moderation Pipeline Integrating Visual Classification, Object Detection, OCR, and Contextual Reasoning for Child Safety
topic Computer Vision and Pattern Recognition
Cryptography and Security
I.4.9; I.2.7; K.4.1
url https://arxiv.org/abs/2603.16181