# Dynamic Content Moderation in Livestreams: Combining Supervised Classification with MLLM-Boosted Similarity Matching

arXiv: https://arxiv.org/abs/2512.03553 · HTML: https://mlsanity.com/paper/2512.03553/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 40 references that are in the map.
- First posted on arXiv 3 Dec 2025.

## Builds on (most cited first, 25 of 25)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks](https://mlsanity.com/paper/2312.14238/) (2023)
- [An Empirical Study of Training End-to-End Vision-and-Language Transformers](https://mlsanity.com/paper/2111.02387/) (2021)
- [ShieldGemma: Generative AI Content Moderation Based on Gemma](https://mlsanity.com/paper/2407.21772/) (2024)
- [HateBERT: Retraining BERT for Abusive Language Detection in English](https://mlsanity.com/paper/2010.12472/) (2020)
- [Contrastive Distillation on Intermediate Representations for Language Model Compression](https://mlsanity.com/paper/2009.14167/) (2020)
- [Policy-as-Prompt: Rethinking Content Moderation in the Age of Large Language Models](https://mlsanity.com/paper/2502.18695/) (2025)
- [VLM as Policy: Common-Law Content Moderation Framework for Short Video Platform](https://mlsanity.com/paper/2504.14904/) (2025)
- [Conceptualising Contestability: Perspectives on Contesting Algorithmic Decisions](https://mlsanity.com/paper/2103.01774/) (2021)
- [TransVCL: Attention-enhanced Video Copy Localization Network with Flexible Supervision](https://mlsanity.com/paper/2211.13090/) (2022)
- [Predictive Embeddings for Hate Speech Detection on Twitter](https://mlsanity.com/paper/1809.10644/) (2018)
- [Enhancing Knowledge Distillation of Large Language Models through Efficient Multi-Modal Distribution Alignment](https://mlsanity.com/paper/2409.12545/) (2024)
- [Advancing Content Moderation: Evaluating Large Language Models for Detecting Sensitive Content Across Text, Images, and Videos](https://mlsanity.com/paper/2411.17123/) (2024)
- [Learning Segment Similarity and Alignment in Large-Scale Content Based Video Retrieval](https://mlsanity.com/paper/2309.11091/) (2023)
- [A Multimodal CNN-based Tool to Censure Inappropriate Video Scenes](https://mlsanity.com/paper/1911.03974/) (2019)
- [Info-Coevolution: An Efficient Framework for Data Model Coevolution](https://mlsanity.com/paper/2506.08070/) (2025)
- [Measuring the Mental Health of Content Reviewers, a Systematic Review](https://mlsanity.com/paper/2502.00244/) (2025)

## Built on by (most cited first, 4 of 4)

- [MatchLM2Lite: A Scalable MLLM-to-Lite Framework for Reproduced Content Identification](https://mlsanity.com/paper/2606.14786/) (2026)
- [MLT-Dedup: Efficient Large-Scale Online Video Deduplication via Multi-Level Representations and Spatial-Temporal Matching](https://mlsanity.com/paper/2606.12215/) (2026)
- [PolicyMem: Geometric Policy Memory for LLM Governance](https://mlsanity.com/paper/2609.13734/) (2026)
- [Live Assistant: Learning Whether, When, and Whom to Assist in Real-World Live Social Streams](https://mlsanity.com/paper/2609.27303/) (2026)

## Abstract (arXiv)

Content moderation remains a critical yet challenging task for large-scale user-generated video platforms, especially in livestreaming environments where moderation must be timely, multimodal, and robust to evolving forms of unwanted content. We present a hybrid moderation framework deployed at production scale that combines supervised classification for known violations with reference-based similarity matching for novel or subtle cases. This hybrid design enables robust detection of both explicit violations and novel edge cases that evade traditional classifiers. Multimodal inputs (text, audio, visual) are processed through both pipelines, with a multimodal large language model (MLLM) distilling knowledge into each to boost accuracy while keeping inference lightweight. In production, the classification pipeline achieves 67% recall at 80% precision, and the similarity pipeline achieves 76% recall at 80% precision. Large-scale A/B tests show a 6-8% reduction in user views of unwanted livestreams}. These results demonstrate a scalable and adaptable approach to multimodal content governance, capable of addressing both explicit violations and emerging adversarial behaviors.