Reasoning-Enhanced Domain-Adaptive Pretraining of Multimodal Large Language Models for Short Video Content Moderation
Zixuan Wang, Yu Sun · 2025
Read Reasoning-Enhanced Domain-Adaptive Pretraining of Multimodal Large Language Models for Short Video Content Moderation by Zixuan Wang, Yu Sun (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-21486.txt.
Description
Short video platforms need one system that can flag many kinds of inappropriate content. Training a small classifier per issue needs much labelled data and generalizes poorly. Zixuan Wang and 8 co-authors propose a unified multimodal approach in a 2025 paper. Domain smart pre training reshapes a large multimodal model for short video governance. Three targeted pre training tasks drive the adaptation.
Caption tasks sharpen perception of video details. Visual question answering deepens understanding of issue definitions and annotation guidelines. Chain of thought tasks strengthen thinking about hard cases. A distribution gap separates short video content from first pre training data, and complex definitions add difficulty. Tests show the paradigm improves unified detection across issue types, with thinking as a core part.
Catalog record
| CREATOR | Zixuan Wang, Yu Sun |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 28 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-21486