Exploring Explanations Improves the Robustness of In-Context Learning

Computer Science

• Large Language Model 100%
• Experimental Result 50%
• Data Distribution 50%
• Decision-Making 50%
• Learning Approach 50%
• Reliability Prediction 50%
• Natural Language Understanding 50%

Keyphrases

• In-context Learning 100%
• Large Language Models 25%
• Recent Advancements 12%
• Natural Language Understanding 12%
• Comprehensive Decision-making 12%
• Out-of-distribution Data 12%
• Robust Decision Making 12%
• Prediction Reliability 12%