Datasets:

Modalities:
Tabular
Text
Formats:
parquet
Languages:
English
ArXiv:
DOI:
Libraries:
Datasets
Dask
License:
quants / README.md
felixdivo's picture
Update README.md
0e78849 verified
|
raw
history blame
7.37 kB
---
dataset_info:
- config_name: binary
features:
- name: sample_id
dtype: int32
- name: question_id
dtype: int32
- name: trajectory
dtype:
array3_d:
shape:
- 320
- 24
- 3
dtype: float32
- name: action_sequence
sequence:
- name: start
dtype: float32
- name: end
dtype: float32
- name: action
dtype: string
- name: action_sentence
dtype: string
length: 4
- name: textual_description
dtype: string
- name: question_type
dtype: string
- name: question
dtype: string
- name: answer_type
dtype: string
- name: answer_text
dtype: string
- name: answer
dtype:
class_label:
names:
'0': 'true'
'1': 'false'
splits:
- name: val
num_bytes: 780073603
num_examples: 6235
- name: test
num_bytes: 765035777
num_examples: 6115
- name: train
num_bytes: 6187058124
num_examples: 49454
download_size: 5725425107
dataset_size: 7732167504
- config_name: multi
features:
- name: sample_id
dtype: int32
- name: question_id
dtype: int32
- name: trajectory
dtype:
array3_d:
shape:
- 320
- 24
- 3
dtype: float32
- name: action_sequence
sequence:
- name: start
dtype: float32
- name: end
dtype: float32
- name: action
dtype: string
- name: action_sentence
dtype: string
length: 4
- name: textual_description
dtype: string
- name: question_type
dtype: string
- name: question
dtype: string
- name: answer_type
dtype: string
- name: answer_text
dtype: string
- name: options
struct:
- name: A
dtype: string
- name: B
dtype: string
- name: C
dtype: string
- name: answer
dtype:
class_label:
names:
'0': '0'
'1': '1'
'2': '2'
- name: options_sequence
sequence: string
length: 3
splits:
- name: val
num_bytes: 541952411
num_examples: 4326
- name: test
num_bytes: 544471458
num_examples: 4346
- name: train
num_bytes: 4375578410
num_examples: 34927
download_size: 4048318598
dataset_size: 5462002279
- config_name: open
features:
- name: sample_id
dtype: int32
- name: question_id
dtype: int32
- name: trajectory
dtype:
array3_d:
shape:
- 320
- 24
- 3
dtype: float32
- name: action_sequence
sequence:
- name: start
dtype: float32
- name: end
dtype: float32
- name: action
dtype: string
- name: action_sentence
dtype: string
length: 4
- name: textual_description
dtype: string
- name: question_type
dtype: string
- name: question
dtype: string
- name: answer_type
dtype: string
- name: answer_text
dtype: string
splits:
- name: val
num_bytes: 555326923
num_examples: 4439
- name: test
num_bytes: 567832548
num_examples: 4539
- name: train
num_bytes: 4455975352
num_examples: 35619
download_size: 4140639676
dataset_size: 5579134823
configs:
- config_name: binary
data_files:
- split: val
path: binary/val-*
- split: test
path: binary/test-*
- split: train
path: binary/train-*
- config_name: multi
data_files:
- split: val
path: multi/val-*
- split: test
path: multi/test-*
- split: train
path: multi/train-*
- config_name: open
data_files:
- split: val
path: open/val-*
- split: test
path: open/test-*
- split: train
path: open/train-*
license: cc-by-4.0
task_categories:
- question-answering
language:
- en
pretty_name: QuAnTS
size_categories:
- 100K<n<1M
---
# QuAnTS: Question Answering on Time Series
[![Dataset Generation GitHub Repo](https://img.shields.io/badge/GitHub-Dataset%20Generation-blue?logo=github)](https://github.com/mauricekraus/quants-generate)
[![arXiv](https://img.shields.io/badge/arXiv-2511.05124-b31b1b.svg)](https://arxiv.org/abs/2511.05124)
QuAnTS is a challenging dataset designed to bridge the gap in question-answering research on time series data.
The dataset features a wide variety of questions and answers concerning human movements, presented as tracked skeleton trajectories.
QuAnTS also includes human reference performance to benchmark the practical usability of models trained on this dataset.
<img src="doc/intro-chat.png" alt="Example chat motivating time series question answering: Q: 'What is the person doing first?', A: 'They are waving.', Q: 'How many times are they jumping after that?', A: '...'" width="30%"/>
At present, there is no official leaderboard for this dataset.
## Dataset Generation Overview
![QuAnTS is generated in several steps: An action sequence is sampled ➀, where for each we sample five question and answer types ➁. For diversity, each of them is then instantiated from a sampled template ➂. The time series from the human motion diffusion ➃ is then combined with the QA-pair and auxiliary data ➄. Example QA pairs are shown below. Dice indicate randomized operations for dataset diversity.](doc/overview.png "Dataset generation overview")
For details, please refer to the paper: *Under Review*
## Task and Format
The primary task for the QuAnTS dataset is Time Series Question Answering.
Given a time series of human skeleton trajectories and a question in natural language, the goal is to generate a correct answer.
Answers are provided in one of the following formats: binary (Yes/No), multiple-choice (A/B/C), or open (free text).
Additionally, to provide more training data for free-text answers, we provide entirely textual answers for all binary and multiple-choice questions.
The ground truth action sequence or scene descriptions *may not* be used to answer the dataset — we provide them for debugging purposes only.
The text in the dataset is in English.
We provide fixed splits into training, validation, and test portions, where only the latter may be used to compare performance across different approaches.
You are free to mix the training and validation splits as needed.
## Licensing, Citation, and Acknowledgments
The QuAnTS dataset is licensed under the [Creative Commons Attribution 4.0 International (CC BY 4.0)](ttps://creativecommons.org/licenses/by/4.0/ ) license.
If you use the QuAnTS dataset in your research, please cite [the paper]([2511.05124](https://arxiv.org/abs/2511.05124)):
```
@misc{divo2025quantsquestionansweringtime,
title={QuAnTS: Question Answering on Time Series},
author={Felix Divo and Maurice Kraus and Anh Q. Nguyen and Hao Xue and Imran Razzak and Flora D. Salim and Kristian Kersting and Devendra Singh Dhami},
year={2025},
eprint={2511.05124},
archivePrefix={arXiv},
primaryClass={cs.LG},
url={https://arxiv.org/abs/2511.05124},
}
```
The dataset was curated by a team of researchers from various institutions:
* Felix Divo, Maurice Kraus, and Kristian Kersting (hessian.AI, DFKI, and the Centre for Cognitive Science) from Technische Universität Darmstadt.
* Anh Q. Nguyen, Hao Xue, and Flora D. Salim from UNSW Sydney.
* Imran Razzak from Mohamed bin Zayed University of Artificial Intelligence.
* Devendra Singh Dhami from Eindhoven University of Technology.