CALM-Bench: A Multi-task Benchmark for Evaluating Causality-Aware Language Models

33427

Knowledge Triples

Source Entity	Relationship	Target Entity
CALM Bench Material	evaluates	CausalQA Material
CALM Bench Material	evaluates	Foundation Language Models OtherEntity
Foundation Language Models OtherEntity	includes	BERT Method
Foundation Language Models OtherEntity	includes	ERNIE Method
Foundation Language Models OtherEntity	includes	T5 Method
Question Answering Task	includes	Abductive Reasoning Task
Question Answering Task	includes	CommonsenseQA Task
Question Answering Task	includes	Procedural Reasoning Task