Adversarial Vulnerabilities in Large Language Models for Time Series Forecasting

Mathematics

• Time Series Forecasting 100%
• Large Language Model 100%
• Adversarial Attack 50%
• Black Box 16%
• Random Noise 16%
• forecasting accuracy 16%

Computer Science

• Large Language Model 100%
• Adversarial Machine Learning 50%
• Temporal Data 16%
• World Application 16%
• Model Architecture 16%
• Code Repository 16%
• forecasting accuracy 16%
• Performance Degradation 16%
• Generative Pre-Trained Transformer 4 16%
• Critical Vulnerability 16%

Keyphrases

• Adversarial Vulnerability 100%
• TimeGPT 16%