Datasets:

Size Categories:
n<1K
ArXiv:
Tags:
lost-in-the-middle
License:
The dataset viewer is taking too long to fetch the data. Try to refresh this page.
Server-side error
Error code:   ClientConnectionError

Datasets for Lost In The Middle

This repository contains datasets used in the paper "Lost in the Middle: How Language Models Use Long Contexts", focusing on multi-document question answering and key-value retrieval tasks.

Datasets Overview

The datasets provided are as follows:

  • Key-Value Retrieval Datasets

    • kv75: Key-Value pairs with 75 keys.
    • kv140: Key-Value pairs with 140 keys.
    • kv300: Key-Value pairs with 300 keys.
  • Multi-Document Question Answering Datasets

    • qa10: Questions with answers derived from 10 documents.
    • qa20: Questions with answers derived from 20 documents.
    • qa30: Questions with answers derived from 30 documents.

Loading the Data

You can load these datasets using the Hugging Face datasets library:

from datasets import load_dataset

### Example for loading the kv75 dataset
dataset = load_dataset("bzantium/LITM", "kv75")

### Example for loading the qa20 dataset
dataset = load_dataset("bzantium/LITM", "qa20")
Downloads last month
26
Edit dataset card