The DOI matches its Crossref record.
Enhanced Multimodal Content Moderation of Children’s Videos using Audiovisual Fusion
Open accessAbstract
Due to the rise in video content creation targeted towards children, there is a need for robust content moderation schemes for video hosting platforms. A video that is visually benign may include audio content that is inappropriate for young children while being impossible to detect with a unimodal content moderation system. Popular video hosting platforms for children such as YouTube Kids still publish videos which contain audio content that is not conducive to a child's healthy behavioral and physical development. A robust classification of malicious videos requires audio representations in addition to video features. However, recent content moderation approaches rarely employ multimodal architectures that explicitly consider non-speech audio cues. To address this, we present an efficient adaptation of CLIP (Contrastive Language–Image Pre-training) that can leverage contextual audio cues for enhanced content moderation. We incorporate 1) the audio modality and 2) prompt learning, while keeping the backbone modules of each modality frozen. We conduct our experiments on a multimodal version of the MOB (Malicious or Benign) dataset in supervised and few-shot settings.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“These capabilities are increasingly vital for agents operating in domains such as robotics [88], autonomous vehicles [89,90], and visual content moderation [91,92].”
From AI Agents vs. Agentic AI: A Conceptual taxonomy, applications and challenges (2026) · cites this paper as [92]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
No references could be resolved yet.
Research works printed but not linked: 24
- Ahmed, S. H.; Khan, M. J.; Qaisar, H. M. U.; and Sukthankar, G. 2023. Malicious or benign? Towards effective content moderation for children's videos. The International FLAIRS Conference Proceedings 36(1).
- Alghowinem, S. 2019. A safer YouTube Kids: An extra layer of content filtering using automated multimodal analysis. In Arai, K.; Kapoor, S.; and Bhatia, R., eds., Intelligent Systems and Applications, 294-308. Cham: Springer International Publishing.
- Cao, T.; Law, M.; and Fidler, S. 2019. A theoretical analysis of the number of shots in few- shot learning. arXiv preprint arXiv:1909.11722. · arXiv
- Chuttur, M. Y., and Nazurally, A. 2022. A multi-modal approach to detect inappropriate cartoon video contents using deep learning networks. Mul- timedia Tools and Applications 81(12):16881-16900. [COPPA 1998] COPPA. 1998. Children's online privacy protection act of 1998. Public Law 105-277, 112 Stat. 2681- 728.
- Deng, J.; Dong, W.; Socher, R.; Li, L.-J.; Li, K.; and Fei-Fei, L. 2009. ImageNet: A large-scale hier- archical image database. In IEEE Conference on Computer Vision and Pattern Recognition, 248-255.
- Elizalde, B.; Deshmukh, S.; Ismail, M. A.; and Wang, H. 2022. CLAP: Learning Audio Concepts From Natural Language Supervision. arXiv:2206.04769 [cs, eess]. · arXiv
- Gemmeke, J. F.; Ellis, D. P.; Freed- man, D.; Jansen, A.; Lawrence, W.; Moore, R. C.; Plakal, M.; and Ritter, M. 2017. Audio set: An ontology and human-labeled dataset for audio events. In IEEE Interna- tional Conference on Acoustics, Speech and Signal Process- ing (ICASSP), 776-780. IEEE.
- Guzhov et al. 2021a] Guzhov, A.; Raue, F.; Hees, J.; and Dengel, A. 2021a. Esresne (x) t-fbsp: Learning robust time- frequency transformation of audio. In International Joint Conference on Neural Networks (IJCNN), 1-8. IEEE.
- Hapkiewicz, W. G. 1979. Children's re- actions to cartoon violence. Journal of Clinical Child Psy- chology 8(1):30-34.
- Hinkley, T., and McCann, J. R. 2018. Mothers' and father's perceptions of the risks and ben- efits of screen time and physical activity during early child- hood: a qualitative study. BMC Public Health 18(1):1-8.
- Khattak, M. U.; Rasheed, H.; Maaz, M.; Khan, S.; and Khan, F. S. 2023. MaPLe: Multi-modal prompt learning. (arXiv:2210.03117). arXiv:2210.03117 [cs]. · arXiv
- Klatte, M.; Bergström, K.; and Lachmann, T. 2013. Does noise affect learning? A short review on noise effects on cognitive performance in children. Frontiers in Psychology 4.
- Le, B. M.; Tandon, R.; and Oinar, C. 2022. Samba: Identifying inappropriate videos for young children on YouTube. In Proceedings of the ACM International Conference on Information and Knowledge Management.
- Lillard, A. S., and Peterson, J. 2011. The immediate impact of different types of tele- vision on young children's executive function. Pediatrics 128(4):644-649.
- McLoughlin, I.; Zhang, H.; Xie, Z.; Song, Y.; and Xiao, W. 2015. Robust sound event classification using deep neural networks. IEEE/ACM Transactions on Audio, Speech, and Language Processing 23(3):540-552.
- Niizumi, D.; Takeuchi, D.; Ohishi, Y.; Harada, N.; and Kashino, K. 2023. BYOL for Audio: Exploring Pre-Trained General-Purpose Audio Representa- tions. IEEE/ACM Transactions on Audio, Speech, and Lan- guage Processing 31:137-151.
- Papadamou, K.; Papasavva, A.; Zannettou, S.; Blackburn, J.; Kourtellis, N.; Leontiadis, I.; Stringhini, G.; and Sirivianos, M. 2021. Disturbed YouTube for Kids: Characterizing and detecting inappropri- ate videos targeting young children. (arXiv:1901.07046). arXiv:1901.07046 [cs]. · arXiv
- Piczak, K. J. 2015. ESC: Dataset for Envi- ronmental Sound Classification. In Proceedings of the 23rd Annual ACM Conference on Multimedia, 1015-1018. ACM Press.
- Rasheed, H.; Khattak, M. U.; Maaz, M.; Khan, S.; and Khan, F. S. 2023. ViFi-CLIP: Fine-tuned CLIP models are efficient video learners. (arXiv:2212.03640). arXiv:2212.03640 [cs]. · arXiv
- Shin, T.; Razeghi, Y.; Logan IV, R. L.; Wallace, E.; and Singh, S. 2020. AutoPrompt: Elicit- ing Knowledge from Language Models with Automatically Generated Prompts. In Webber, B.; Cohn, T.; He, Y.; and Liu, Y., eds., Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), 4222- 4235. Online: Association for Computational Linguistics.
- Vimal, B.; Surya, M.; Darshan; Sridhar, V.; and Ashok, A. 2021. MFCC based audio classifica- tion using machine learning. In International Conference on Computing Communication and Networking Technolo- gies (ICCCNT), 1-4.
- Wasim, S. T.; Naseer, M.; Khan, S.; Khan, F. S.; and Shah, M. 2023. Vita-CLIP: Video and text adaptive CLIP via multimodal prompting. (arXiv:2304.03307). arXiv:2304.03307 [cs, eess]. [YouTube 2024] YouTube. 2024. How does YouTube man- age harmful content? [ · arXiv
- Zhou et al. 2022a] Zhou, K.; Yang, J.; Loy, C. C.; and Liu, Z. 2022a. Cocoop: Conditional prompt learning for vision- language models. (arXiv:2203.05557). · arXiv
- Zhou et al. 2022b] Zhou, K.; Yang, J.; Loy, C. C.; and Liu, Z. 2022b. Coop: Learning to prompt for vision- language models. International Journal of Computer Vision 130(9):2337-2348.
Cited by
Sorted by citations
AI Agents vs. Agentic AI: A Conceptual taxonomy, applications and challenges
Sapkota, R., Roumeliotis, K. I., & Karkee, M. (2026) · Information Fusion · doi:10.1016/j.inffus.2025.103599 · Cited by 671
Metadata
- Authors
- Ahmed, S. H., Khan, M. J., & Sukthankar, G.
- Affiliations
- University of Central Florida (US)
- Venue
- The International FLAIRS Conference Proceedings
- Published
- 2024-05-13
- Volume
- 37
- ISSN
- 2334-0754 (ISSN-L)
- 2334-0762
- OpenAlex
- W4400348305
- Cited by
- 4
- References
- 0
- Citations by year
- 2026: 2 · 2025: 1 · 2024: 1
- FWCI
- 1.15
- Type
- conference paper
- Language
- en
- Open access
- diamond
- doi.org
- Licence
- Locations
- Proceedings of the ... International Florida Artificial Intelligence Research Society Conference (publishedVersion, cc-by-nc)
- arXiv (Cornell University) (submittedVersion, cc-by-nc-sa)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Hate Speech and Cyberbullying Detection
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Multimodal Machine Learning Applications
- Physical Sciences › Computer Science › Signal Processing › Music and Audio Processing
- Keywords
- audiovisual fusion, prompt learning, few-shot learning, YouTube Kids
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
