Dataset Viewer
source
stringclasses 2
values | split
stringclasses 2
values | image_id
stringlengths 10
20
|
---|---|---|
docci
|
test
|
test_00000
|
docci
|
test
|
test_00001
|
docci
|
test
|
test_00002
|
docci
|
test
|
test_00003
|
docci
|
test
|
test_00004
|
docci
|
test
|
test_00005
|
docci
|
test
|
test_00006
|
docci
|
test
|
test_00007
|
docci
|
test
|
test_00008
|
docci
|
test
|
test_00009
|
docci
|
test
|
test_00010
|
docci
|
test
|
test_00011
|
docci
|
test
|
test_00012
|
docci
|
test
|
test_00013
|
docci
|
test
|
test_00014
|
docci
|
test
|
test_00015
|
docci
|
test
|
test_00016
|
docci
|
test
|
test_00017
|
docci
|
test
|
test_00018
|
docci
|
test
|
test_00019
|
docci
|
test
|
test_00020
|
docci
|
test
|
test_00021
|
docci
|
test
|
test_00022
|
docci
|
test
|
test_00023
|
docci
|
test
|
test_00024
|
docci
|
test
|
test_00025
|
docci
|
test
|
test_00026
|
docci
|
test
|
test_00027
|
docci
|
test
|
test_00028
|
docci
|
test
|
test_00029
|
docci
|
test
|
test_00030
|
docci
|
test
|
test_00031
|
docci
|
test
|
test_00032
|
docci
|
test
|
test_00033
|
docci
|
test
|
test_00034
|
docci
|
test
|
test_00035
|
docci
|
test
|
test_00036
|
docci
|
test
|
test_00037
|
docci
|
test
|
test_00038
|
docci
|
test
|
test_00039
|
docci
|
test
|
test_00040
|
docci
|
test
|
test_00041
|
docci
|
test
|
test_00042
|
docci
|
test
|
test_00043
|
docci
|
test
|
test_00044
|
docci
|
test
|
test_00045
|
docci
|
test
|
test_00046
|
docci
|
test
|
test_00047
|
docci
|
test
|
test_00048
|
docci
|
test
|
test_00049
|
docci
|
test
|
test_00050
|
docci
|
test
|
test_00051
|
docci
|
test
|
test_00052
|
docci
|
test
|
test_00053
|
docci
|
test
|
test_00054
|
docci
|
test
|
test_00055
|
docci
|
test
|
test_00056
|
docci
|
test
|
test_00057
|
docci
|
test
|
test_00058
|
docci
|
test
|
test_00059
|
docci
|
test
|
test_00060
|
docci
|
test
|
test_00061
|
docci
|
test
|
test_00062
|
docci
|
test
|
test_00063
|
docci
|
test
|
test_00064
|
docci
|
test
|
test_00065
|
docci
|
test
|
test_00066
|
docci
|
test
|
test_00067
|
docci
|
test
|
test_00068
|
docci
|
test
|
test_00069
|
docci
|
test
|
test_00070
|
docci
|
test
|
test_00071
|
docci
|
test
|
test_00072
|
docci
|
test
|
test_00073
|
docci
|
test
|
test_00074
|
docci
|
test
|
test_00075
|
docci
|
test
|
test_00076
|
docci
|
test
|
test_00077
|
docci
|
test
|
test_00078
|
docci
|
test
|
test_00079
|
docci
|
test
|
test_00080
|
docci
|
test
|
test_00081
|
docci
|
test
|
test_00082
|
docci
|
test
|
test_00083
|
docci
|
test
|
test_00084
|
docci
|
test
|
test_00085
|
docci
|
test
|
test_00086
|
docci
|
test
|
test_00087
|
docci
|
test
|
test_00088
|
docci
|
test
|
test_00089
|
docci
|
test
|
test_00090
|
docci
|
test
|
test_00091
|
docci
|
test
|
test_00092
|
docci
|
test
|
test_00093
|
docci
|
test
|
test_00094
|
docci
|
test
|
test_00095
|
docci
|
test
|
test_00096
|
docci
|
test
|
test_00097
|
docci
|
test
|
test_00098
|
docci
|
test
|
test_00099
|
End of preview. Expand
in Data Studio
ND400 Derivative Dataset
Description:
ND400 is a derivative evaluation dataset created to study semantic drift in Unified Models (UMs) across multiple image-text generations. It contains 400 samples:
- 200 from DOCCI test set (
test_00000
–test_00199
) - 200 from NoCaps validation set (first 200 entries)
This dataset does not redistribute original images. It provides a CSV of image IDs, splits, and sources to enable users to fetch the images from the original datasets. A sample script to fetch the original dataset will be provided in the project GitHub repo.
CSV Columns:
source | split | image_id |
---|---|---|
docci | test | test_00000 |
nocaps | validation | 000123.jpg |
Usage:
- Download the CSV (
nd400.csv
) from this dataset. - Use the provided prepare_dataset.py to organize images and captions for evaluation:
<evaluation_data_root>/
├── text-first/
│ ├── gen-0.csv
├── image-first/
│ └── gen-0/ (images)
License
This dataset combines data from DOCCl (CC BY 4.0) and NoCaps (see original). Please respect the original licenses.
Citation:
@misc{mollah2025telephonegameevaluatingsemantic,
title={The Telephone Game: Evaluating Semantic Drift in Unified Models},
author={Sabbir Mollah and Rohit Gupta and Sirnam Swetha and Qingyang Liu and Ahnaf Munir and Mubarak Shah},
year={2025},
eprint={2509.04438},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://arxiv.org/abs/2509.04438},
}
- Downloads last month
- 58