Feature Request: Commonsense Reasoning - Build a Model for Common-Sense Inference

[HsiangNianian]

Commonsense reasoning involves reasoning with everyday knowledge that humans use to make sense of the world. This task will involve building models that can understand and apply common-sense knowledge in real-world scenarios.

• Datasets: What commonsense reasoning datasets should we use (e.g., ATOMIC, SWAG)?
• Model Choice: Should we use models like GPT-3, BERT, or T5 for commonsense tasks?
• Application: How can we apply commonsense reasoning to real-world applications (e.g., chatbots, reasoning-based question answering)?

Expected Outcome:

• A commonsense reasoning model capable of understanding everyday knowledge and inferring logical conclusions.
• Integration with iamai for better conversational agents and automated reasoning tasks.

Labels: feature, reasoning, commonsense