Summary of Detection Of Conspiracy Theories Beyond Keyword Bias in German-language Telegram Using Large Language Models, by Milena Pustet et al.
Detection of Conspiracy Theories Beyond Keyword Bias in German-Language Telegram Using Large Language Models
by Milena Pustet, Elisabeth Steffen, Helena Mihaljević
First submitted to arxiv on: 27 Apr 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Social and Information Networks (cs.SI)
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 an innovative approach to detecting conspiracy theories online by comparing the performance of supervised fine-tuning and prompt-based methods in detecting German Telegram messages. The study focuses on developing models that require minimal additional training data and addresses the issue of token-level bias introduced by available datasets, which are predominantly in English. By leveraging BERT-like models and Llama2, GPT-3.5, and GPT-4 prompts, the authors aim to improve the accuracy of conspiracy theory detection and provide insights into the effectiveness of different approaches. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper explores ways to detect conspiracy theories online without needing a lot of training data. The researchers compared two types of models: ones that learn from labeled examples (supervised) and ones that can be guided with specific questions or prompts. They tested these models on a dataset of over 4,000 German-language messages shared during the COVID-19 pandemic, without using any filters to focus on specific keywords. The goal is to develop better ways to spot conspiracy theories online and make the internet a safer place. |
Keywords
» Artificial intelligence » Bert » Fine tuning » Gpt » Prompt » Supervised » Token
