TriviaQA

Introduced by Joshi et al. in TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension

TriviaQA is a realistic text-based question answering dataset which includes 950K question-answer pairs from 662K documents collected from Wikipedia and the web. This dataset is more challenging than standard QA benchmark datasets such as Stanford Question Answering Dataset (SQuAD), as the answers for a question may not be directly obtained by span prediction and the context is very long. TriviaQA dataset consists of both human-verified and machine-generated QA subsets.

Source: Episodic Memory Reader: Learning What to Rememberfor Question Answering from Streaming Data

Homepage

Benchmarks

Add a new result Link an existing benchmark

Task	Dataset Variant	Best Model
Question Answering	TriviaQA	Claude 2
Open-Domain Question Answering	KILT: TriviaQA	Re2G
Question Generation	TriviaQA	Info-HCVAE
Text Generation	TriviaQA	bloom
Open-Domain Question Answering	TriviaQA	UnitedQA