Iker/ClickbaitFighter-2B
Iker/ClickbaitFighter-2B is a 2.6 billion parameter causal language model developed by Iker, fine-tuned from Google's Gemma-2B-IT. Optimized for Spanish, this model specializes in debunking sensationalist or clickbait headlines by generating concise, factual summaries. It achieves a ROUGE score of 36.26 on the NoticIA dataset, making it suitable for automated clickbait detection and summarization tasks.
Loading preview...
Iker/ClickbaitFighter-2B: Spanish Clickbait Summarization Model
Iker/ClickbaitFighter-2B is a 2.6 billion parameter language model, fine-tuned from google/gemma-2b-it, specifically designed for the task of identifying and summarizing clickbait headlines in Spanish. This model was developed by Iker and trained using the NoticIA Dataset, a specialized dataset for clickbait article summarization in Spanish.
Key Capabilities
- Clickbait Debunking: Analyzes sensationalist or clickbait headlines and generates a single-sentence summary that reveals the truth behind the headline.
- Contextual Summarization: Extracts relevant information from the article body to provide a factual and concise response, often quoting original text.
- Spanish Language Focus: Optimized for Spanish content, making it highly effective for Spanish news analysis.
- Performance: Achieves a ROUGE score of 36.26, demonstrating its ability to generate relevant summaries.
Good for
- Automated News Analysis: Ideal for applications requiring the automatic identification and summarization of clickbait content in Spanish.
- Content Moderation: Can assist in flagging and rephrasing misleading headlines.
- Information Extraction: Useful for extracting core facts from articles with sensationalist titles.
For higher performance, users are encouraged to explore larger models in the ClickbaitFighter series, such as ClickbaitFighter-10B.