Improving Text Embeddings with Large Language Models
Paper
• 2401.00368 • Published
• 82
query stringclasses 50 values | candidates sequencelengths 100 100 | label int64 0 0 | context_length int64 256 32.8k |
|---|---|---|---|
what is the passkey for Cohen Figueroa? | [
"The grass is green. The sky is blue. The sun is yellow. Here we go. There and back again. The grass is green. The sky is blue. The sun is yellow. Here we go. There and back again. The grass is green. The sky is blue. The sun is yellow. Here we go. There and back again. The grass is green. The sky is blue. The \nCo... | 0 | 256 |
what is the passkey for Eliel McKee? | ["The grass is green. The sky is blue. The sun is yellow. Here we go. There and back again. The gras(...TRUNCATED) | 0 | 256 |
what is the passkey for Ace Nichols? | ["The grass is green. The sky is blue. The sun is yellow. Here we go. There and back again. The gras(...TRUNCATED) | 0 | 256 |
what is the passkey for Aileen Munoz? | ["The grass is green. The sky is blue. The sun is yellow. Here we go. There and back again. The gras(...TRUNCATED) | 0 | 256 |
what is the passkey for Vivian Villalobos? | ["The grass is green. The sky is blue. The sun is yellow. Here we go. There and back again. The gras(...TRUNCATED) | 0 | 256 |
what is the passkey for Sarai Malone? | ["The grass is green. The sky is blue. The sun is yellow. Here we go. There and back again. The gras(...TRUNCATED) | 0 | 256 |
what is the passkey for Aaliyah Shepherd? | ["The grass is green. The sky is blue. The sun is yellow. Here we go. There and back again. The gras(...TRUNCATED) | 0 | 256 |
what is the passkey for Kailani Waters? | ["The grass is green. The sky is blue. The sun is yellow. Here we go. There and back again. The gras(...TRUNCATED) | 0 | 256 |
what is the passkey for Alaia Walker? | ["T\nAlaia Walker's pass key is 2525. Remember it. 2525 is the pass key for Alaia Walker.\nThe grass(...TRUNCATED) | 0 | 256 |
what is the passkey for Sky Morrow? | ["The grass is green. The sky is blue. The sun is yellow. Here we go. There and back again. The gras(...TRUNCATED) | 0 | 256 |
This dataset contains the data for personalized passkey retrieval task in the paper Improving Text Embeddings with Large Language Models.
query: a string feature.candidates: List of string feature, 100 candidates for each query.label: a int32 feature, the index of the correct candidate in the candidates list, always 0.context_length: a int32 feature, the approximate length for the candidate documents.You can load the dataset in your python code as follows:
from datasets import load_dataset
dataset = load_dataset("intfloat/personalized_passkey_retrieval")
The data in this repo is generated by the script generate_passkey_data.py. You can also tweak the script to generate your own data.
If you use this dataset in your research, please cite this paper:
@inproceedings{Wang2023ImprovingTE,
title={Improving Text Embeddings with Large Language Models},
author={Liang Wang and Nan Yang and Xiaolong Huang and Linjun Yang and Rangan Majumder and Furu Wei},
year={2023},
}