Question 1

What are Text Retrieval datasets used for?

Accepted Answer

Text Retrieval datasets are collections of labelled or raw data used to train, fine-tune, and evaluate models on the text retrieval task. This page lists 6 such datasets, each linking to its source and paper.

Question 2

Which Text Retrieval dataset is best for benchmarking?

Accepted Answer

None of the listed Text Retrieval datasets are currently tracked as standard LLM benchmarks, but many are widely used for evaluation.

Question 3

How many Text Retrieval datasets are there?

Accepted Answer

We catalog 6 Text Retrieval datasets in one searchable directory.

Text Retrieval Datasets

What languages do text retrieval datasets cover?

Explore other dataset tasks

Frequently asked questions