ComplexWebQuestions: A comprehensive dataset designed for tackling complex questions necessitating reasoning across multiple web snippets.
Versatile Usage:
Interacting with Search Engines: The primary focus of our paper (Talmor and Berant, 2018), facilitating exploration and experimentation with search engine interaction.
Reading Comprehension Task: Access to a vast collection of 12,725,989 web snippets relevant to the questions, gathered during our model's development phase.
Semantic Parsing Task: Each question is meticulously paired with a corresponding SPARQL query, enabling execution against Freebase to extract answers efficiently.
