Summary of Comment-aided Video-language Alignment Via Contrastive Pre-training For Short-form Video Humor Detection, by Yang Liu et al.
Comment-aided Video-Language Alignment via Contrastive Pre-training for Short-form Video Humor Detection
by Yang Liu, Tongfei Shen, Dong Zhang, Qingying Sun, Shoushan Li, Guodong Zhou
First submitted to arxiv on: 14 Feb 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed Comment-aided Video-Language Alignment (CVLA) model achieves state-of-the-art results in short-form video humor detection, outperforming multiple baseline approaches and existing methods. This hierarchical two-branch model utilizes data-augmented multi-modal contrastive pre-training to align video and language components within a consistent semantic space. By operating on raw signals across various modal channels, CVLA demonstrates its effectiveness on two humor detection datasets: DY11k and UR-FUNNY. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper proposes a new way to detect funny moments in short videos on social media platforms like TikTok or YouTube. The model uses both the video and text (like captions) to figure out what’s making people laugh. It works really well, beating other methods at this task. This is important because humor is a big part of how we connect with each other online. |
Keywords
* Artificial intelligence * Alignment * Multi modal
