Summary of Mmoe: Robust Spoiler Detection with Multi-modal Information and Domain-aware Mixture-of-experts, by Zinan Zeng et al.
MMoE: Robust Spoiler Detection with Multi-modal Information and Domain-aware Mixture-of-Experts
by Zinan Zeng, Sen Ye, Zijian Cai, Heng Wang, Yuhan Liu, Haokai Zhang, Minnan Luo
First submitted to arxiv on: 8 Mar 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: None
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 In this paper, researchers propose a multi-modal network called MMoE for detecting spoilers in online movie reviews. The method leverages information from multiple modalities, including text, metadata, and graph features from user-movie networks. To address the challenge of genre-specific spoilers, MMoE uses a Mixture-of-Experts architecture to process information in three modalities and promote robustness. Experimental results demonstrate that MMoE achieves state-of-the-art performance on two widely-used spoiler detection datasets, surpassing previous methods by 2.56% and 8.41% in terms of accuracy and F1-score. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper develops a new approach for detecting spoilers in online movie reviews. The method combines information from different sources, like the text of the review, the user’s profile, and the connection between users and movies. This helps the algorithm understand how different genres affect the way spoilers are written. The proposed model is tested on two large datasets and performs better than previous methods. |
Keywords
» Artificial intelligence » F1 score » Mixture of experts » Multi modal
