VLM as Policy: Common-Law Content Moderation Framework for Short Video Platform

2025-04-21

AI Summary

Short-video platforms face three critical bottlenecks in moderating harmful content threatening minors’ mental health: (1) high bias and cost of manual review; (2) coarse-grained understanding and low accuracy of automated methods; and (3) regulatory lag, hindering timely adaptation to evolving risks. To address these, we propose KuaiMod—a precedent-driven dynamic moderation framework that pioneers a common-law-inspired paradigm. It integrates large vision-language models (VLMs) with chain-of-thought (CoT) reasoning and enables policy self-adaptation via sparse real-world feedback from users and human reviewers. Our contributions include: (1) the first publicly available benchmark featuring authentic user/reviewer feedback; (2) state-of-the-art performance on this benchmark; (3) a 20% reduction in user-reported violations after online deployment; and (4) significant improvements in multi-scenario DAU and average app session duration upon integration with the recommendation system.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Safety and Robustness

Application Category

• Responsible Web: Mitigating online harms, including harassment, spam, and fake reviews
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Exponentially growing short video platforms (SVPs) face significant challenges in moderating content detrimental to users' mental health, particularly for minors. The dissemination of such content on SVPs can lead to catastrophic societal consequences. Although substantial efforts have been dedicated to moderating such content, existing methods suffer from critical limitations: (1) Manual review is prone to human bias and incurs high operational costs. (2) Automated methods, though efficient, lack nuanced content understanding, resulting in lower accuracy. (3) Industrial moderation regulations struggle to adapt to rapidly evolving trends due to long update cycles. In this paper, we annotate the first SVP content moderation benchmark with authentic user/reviewer feedback to fill the absence of benchmark in this field. Then we evaluate various methods on the benchmark to verify the existence of the aforementioned limitations. We further propose our common-law content moderation framework named KuaiMod to address these challenges. KuaiMod consists of three components: training data construction, offline adaptation, and online deployment&refinement. Leveraging large vision language model (VLM) and Chain-of-Thought (CoT) reasoning, KuaiMod adequately models video toxicity based on sparse user feedback and fosters dynamic moderation policy with rapid update speed and high accuracy. Offline experiments and large-scale online A/B test demonstrates the superiority of KuaiMod: KuaiMod achieves the best moderation performance on our benchmark. The deployment of KuaiMod reduces the user reporting rate by 20% and its application in video recommendation increases both Daily Active User (DAU) and APP Usage Time (AUT) on several Kuaishou scenarios. We have open-sourced our benchmark at https://kuaimod.github.io.

Problem

• Moderate harmful content on short video platforms efficiently
• Overcome biases and high costs in manual content review
• Adapt quickly to evolving trends in content moderation

Innovation

• Uses VLM and CoT for nuanced content understanding
• Implements dynamic policy with rapid updates
• Leverages user feedback for training data