Summary of Crisiskan: Knowledge-infused and Explainable Multimodal Attention Network For Crisis Event Classification, by Shubham Gupta et al.
CrisisKAN: Knowledge-infused and Explainable Multimodal Attention Network for Crisis Event Classification
by Shubham Gupta, Nandini Saini, Suman Kundu, Debasis Das
First submitted to arxiv on: 11 Jan 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
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 This paper proposes CrisisKAN, a novel Knowledge-infused and Explainable Multimodal Attention Network that integrates images, texts, and external knowledge from Wikipedia to classify crisis events. The model aims to bridge the semantic gap between image and text modalities, provide reliable explanations of predictions, and address potential biases in social media posts. To achieve this, CrisisKAN employs a guided cross-attention module, incorporates Wikipedia knowledge using a proposed wiki extraction algorithm, and utilizes Gradient-weighted Class Activation Mapping (Grad-CAM) for robust explanation. The model outperforms existing state-of-the-art methodologies on the CrisisMMD dataset across various crisis-specific tasks and settings. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary CrisisKAN is a new way to use computers to help identify important events happening in real-time, like natural disasters or pandemics. Right now, these events are hard to recognize because images and text can be very different. This paper makes a special computer program that uses both images and texts together with extra information from Wikipedia to better understand what’s happening. The program also tries to explain its own decisions so people can trust it in important situations. |
Keywords
* Artificial intelligence * Attention * Cross attention
