id
stringlengths 3
5
| question
stringlengths 37
82
| options
listlengths 4
4
| answer
stringclasses 9
values | task_plan
stringlengths 69
153
| image
imagewidth (px) 1.02k
1.02k
|
|---|---|---|---|---|---|
0-0 | How many steak are there in the image? | [
"4",
"1",
"2",
"3"
] | 3 | count: 3
grid number: 2
question type: how many
target category: steak | |
0-1 | How many steak are there in the image? | [
"4",
"1",
"2",
"3"
] | 3 | count: 3
grid number: 2
question type: how many
target category: steak | |
0-2 | How many steak are there in the image? | [
"2",
"1",
"3",
"4"
] | 3 | count: 3
grid number: 2
question type: how many
target category: steak | |
1-0 | How many oven are there in the image? | [
"4",
"5",
"9",
"7"
] | 9 | count: 9
grid number: 3
question type: how many
target category: oven | |
1-1 | How many oven are there in the image? | [
"8",
"4",
"9",
"7"
] | 9 | count: 9
grid number: 3
question type: how many
target category: oven | |
1-2 | How many oven are there in the image? | [
"3",
"8",
"4",
"9"
] | 9 | count: 9
grid number: 3
question type: how many
target category: oven | |
2-0 | How many red oven are there in the image? | [
"9",
"1",
"6",
"5"
] | 9 | attribute type: color
attribute value: red
count: 9
grid number: 3
question type: how many
target category: oven | |
2-1 | How many red oven are there in the image? | [
"3",
"9",
"2",
"8"
] | 9 | attribute type: color
attribute value: red
count: 9
grid number: 3
question type: how many
target category: oven | |
2-2 | How many red oven are there in the image? | [
"6",
"9",
"1",
"4"
] | 9 | attribute type: color
attribute value: red
count: 9
grid number: 3
question type: how many
target category: oven | |
3-0 | How many metal oven are there in the image? | [
"5",
"6",
"9",
"8"
] | 9 | attribute type: material
attribute value: metal
count: 9
grid number: 3
question type: how many
target category: oven | |
3-1 | How many metal oven are there in the image? | [
"6",
"3",
"5",
"9"
] | 9 | attribute type: material
attribute value: metal
count: 9
grid number: 3
question type: how many
target category: oven | |
3-2 | How many metal oven are there in the image? | [
"2",
"9",
"5",
"7"
] | 9 | attribute type: material
attribute value: metal
count: 9
grid number: 3
question type: how many
target category: oven | |
4-0 | How many cuboid oven are there in the image? | [
"8",
"9",
"6",
"5"
] | 9 | attribute type: shape
attribute value: cuboid
count: 9
grid number: 3
question type: how many
target category: oven | |
4-1 | How many cuboid oven are there in the image? | [
"9",
"8",
"1",
"3"
] | 9 | attribute type: shape
attribute value: cuboid
count: 9
grid number: 3
question type: how many
target category: oven | |
4-2 | How many cuboid oven are there in the image? | [
"4",
"7",
"9",
"2"
] | 9 | attribute type: shape
attribute value: cuboid
count: 9
grid number: 3
question type: how many
target category: oven | |
5-0 | How many black table lamp are there in the image? | [
"4",
"2",
"3",
"1"
] | 2 | attribute type: color
attribute value: black
count: 2
grid number: 2
question type: how many
target category: table lamp | |
5-1 | How many black table lamp are there in the image? | [
"3",
"4",
"2",
"1"
] | 2 | attribute type: color
attribute value: black
count: 2
grid number: 2
question type: how many
target category: table lamp | |
5-2 | How many black table lamp are there in the image? | [
"1",
"3",
"4",
"2"
] | 2 | attribute type: color
attribute value: black
count: 2
grid number: 2
question type: how many
target category: table lamp | |
6-0 | How many gray table lamp are there in the image? | [
"2",
"1",
"4",
"3"
] | 2 | attribute type: color
attribute value: gray
count: 2
grid number: 2
question type: how many
target category: table lamp | |
6-1 | How many gray table lamp are there in the image? | [
"1",
"4",
"3",
"2"
] | 2 | attribute type: color
attribute value: gray
count: 2
grid number: 2
question type: how many
target category: table lamp | |
6-2 | How many gray table lamp are there in the image? | [
"1",
"4",
"3",
"2"
] | 2 | attribute type: color
attribute value: gray
count: 2
grid number: 2
question type: how many
target category: table lamp | |
7-0 | How many white table lamp are there in the image? | [
"4",
"1",
"3",
"2"
] | 3 | attribute type: color
attribute value: white
count: 3
grid number: 2
question type: how many
target category: table lamp | |
7-1 | How many white table lamp are there in the image? | [
"1",
"3",
"4",
"2"
] | 3 | attribute type: color
attribute value: white
count: 3
grid number: 2
question type: how many
target category: table lamp | |
7-2 | How many white table lamp are there in the image? | [
"3",
"1",
"4",
"2"
] | 3 | attribute type: color
attribute value: white
count: 3
grid number: 2
question type: how many
target category: table lamp | |
8-0 | How many table lamp are there in the image? | [
"1",
"3",
"4",
"2"
] | 4 | count: 4
grid number: 2
question type: how many
target category: table lamp | |
8-1 | How many table lamp are there in the image? | [
"1",
"4",
"3",
"2"
] | 4 | count: 4
grid number: 2
question type: how many
target category: table lamp | |
8-2 | How many table lamp are there in the image? | [
"4",
"2",
"3",
"1"
] | 4 | count: 4
grid number: 2
question type: how many
target category: table lamp | |
9-0 | How many metal table lamp are there in the image? | [
"3",
"6",
"4",
"9"
] | 3 | attribute type: material
attribute value: metal
count: 3
grid number: 3
question type: how many
target category: table lamp | |
9-1 | How many metal table lamp are there in the image? | [
"4",
"5",
"1",
"3"
] | 3 | attribute type: material
attribute value: metal
count: 3
grid number: 3
question type: how many
target category: table lamp | |
9-2 | How many metal table lamp are there in the image? | [
"9",
"2",
"6",
"3"
] | 3 | attribute type: material
attribute value: metal
count: 3
grid number: 3
question type: how many
target category: table lamp | |
10-0 | How many blue objects are there in the image? | [
"3",
"2",
"1",
"4"
] | 1 | attribute type: color
attribute value: blue
count: 1
grid number: 2
question type: how many | |
10-1 | How many blue objects are there in the image? | [
"4",
"2",
"1",
"3"
] | 1 | attribute type: color
attribute value: blue
count: 1
grid number: 2
question type: how many | |
10-2 | How many blue objects are there in the image? | [
"1",
"3",
"2",
"4"
] | 1 | attribute type: color
attribute value: blue
count: 1
grid number: 2
question type: how many | |
11-0 | How many volleyball are there in the image? | [
"3",
"1",
"4",
"2"
] | 3 | count: 3
grid number: 2
question type: how many
target category: volleyball | |
11-1 | How many volleyball are there in the image? | [
"1",
"2",
"4",
"3"
] | 3 | count: 3
grid number: 2
question type: how many
target category: volleyball | |
11-2 | How many volleyball are there in the image? | [
"4",
"1",
"2",
"3"
] | 3 | count: 3
grid number: 2
question type: how many
target category: volleyball | |
12-0 | How many flower are there in the image? | [
"3",
"4",
"1",
"2"
] | 2 | count: 2
grid number: 2
question type: how many
target category: flower | |
12-1 | How many flower are there in the image? | [
"4",
"2",
"3",
"1"
] | 2 | count: 2
grid number: 2
question type: how many
target category: flower | |
12-2 | How many flower are there in the image? | [
"4",
"1",
"3",
"2"
] | 2 | count: 2
grid number: 2
question type: how many
target category: flower | |
13-0 | How many shoe are there in the image? | [
"1",
"3",
"2",
"4"
] | 1 | count: 1
grid number: 2
question type: how many
target category: shoe | |
13-1 | How many shoe are there in the image? | [
"4",
"1",
"3",
"2"
] | 1 | count: 1
grid number: 2
question type: how many
target category: shoe | |
13-2 | How many shoe are there in the image? | [
"2",
"1",
"4",
"3"
] | 1 | count: 1
grid number: 2
question type: how many
target category: shoe | |
14-0 | How many broccoli are there in the image? | [
"4",
"9",
"2",
"7"
] | 2 | count: 2
grid number: 3
question type: how many
target category: broccoli | |
14-1 | How many broccoli are there in the image? | [
"9",
"5",
"2",
"1"
] | 2 | count: 2
grid number: 3
question type: how many
target category: broccoli | |
14-2 | How many broccoli are there in the image? | [
"2",
"7",
"5",
"3"
] | 2 | count: 2
grid number: 3
question type: how many
target category: broccoli | |
15-0 | How many seat are there in the image? | [
"2",
"4",
"3",
"1"
] | 1 | count: 1
grid number: 2
question type: how many
target category: seat | |
15-1 | How many seat are there in the image? | [
"2",
"3",
"4",
"1"
] | 1 | count: 1
grid number: 2
question type: how many
target category: seat | |
15-2 | How many seat are there in the image? | [
"2",
"3",
"4",
"1"
] | 1 | count: 1
grid number: 2
question type: how many
target category: seat | |
16-0 | How many paper towel are there in the image? | [
"4",
"9",
"2",
"1"
] | 2 | count: 2
grid number: 3
question type: how many
target category: paper towel | |
16-1 | How many paper towel are there in the image? | [
"2",
"1",
"9",
"7"
] | 2 | count: 2
grid number: 3
question type: how many
target category: paper towel | |
16-2 | How many paper towel are there in the image? | [
"5",
"2",
"6",
"1"
] | 2 | count: 2
grid number: 3
question type: how many
target category: paper towel | |
17-0 | How many metal flashlight are there in the image? | [
"3",
"9",
"1",
"7"
] | 3 | attribute type: material
attribute value: metal
count: 3
grid number: 3
question type: how many
target category: flashlight | |
17-1 | How many metal flashlight are there in the image? | [
"7",
"5",
"3",
"6"
] | 3 | attribute type: material
attribute value: metal
count: 3
grid number: 3
question type: how many
target category: flashlight | |
17-2 | How many metal flashlight are there in the image? | [
"8",
"5",
"2",
"3"
] | 3 | attribute type: material
attribute value: metal
count: 3
grid number: 3
question type: how many
target category: flashlight | |
18-0 | How many yellow personal protective equipment are there in the image? | [
"8",
"4",
"7",
"1"
] | 8 | attribute type: color
attribute value: yellow
count: 8
grid number: 3
question type: how many
target category: personal protective equipment | |
18-1 | How many yellow personal protective equipment are there in the image? | [
"2",
"5",
"1",
"8"
] | 8 | attribute type: color
attribute value: yellow
count: 8
grid number: 3
question type: how many
target category: personal protective equipment | |
18-2 | How many yellow personal protective equipment are there in the image? | [
"6",
"2",
"9",
"8"
] | 8 | attribute type: color
attribute value: yellow
count: 8
grid number: 3
question type: how many
target category: personal protective equipment | |
19-0 | How many sweatpants are there in the image? | [
"3",
"4",
"2",
"1"
] | 1 | count: 1
grid number: 2
question type: how many
target category: sweatpants | |
19-1 | How many sweatpants are there in the image? | [
"2",
"4",
"3",
"1"
] | 1 | count: 1
grid number: 2
question type: how many
target category: sweatpants | |
19-2 | How many sweatpants are there in the image? | [
"4",
"1",
"3",
"2"
] | 1 | count: 1
grid number: 2
question type: how many
target category: sweatpants | |
20-0 | How many orange winter squash are there in the image? | [
"4",
"1",
"3",
"2"
] | 4 | attribute type: color
attribute value: orange
count: 4
grid number: 2
question type: how many
target category: winter squash | |
20-1 | How many orange winter squash are there in the image? | [
"1",
"2",
"3",
"4"
] | 4 | attribute type: color
attribute value: orange
count: 4
grid number: 2
question type: how many
target category: winter squash | |
20-2 | How many orange winter squash are there in the image? | [
"3",
"1",
"2",
"4"
] | 4 | attribute type: color
attribute value: orange
count: 4
grid number: 2
question type: how many
target category: winter squash | |
21-0 | How many hamper are there in the image? | [
"1",
"4",
"2",
"3"
] | 4 | count: 4
grid number: 2
question type: how many
target category: hamper | |
21-1 | How many hamper are there in the image? | [
"2",
"3",
"4",
"1"
] | 4 | count: 4
grid number: 2
question type: how many
target category: hamper | |
21-2 | How many hamper are there in the image? | [
"1",
"3",
"2",
"4"
] | 4 | count: 4
grid number: 2
question type: how many
target category: hamper | |
22-0 | How many orange orange are there in the image? | [
"7",
"8",
"5",
"6"
] | 6 | attribute type: color
attribute value: orange
count: 6
grid number: 3
question type: how many
target category: orange | |
22-1 | How many orange orange are there in the image? | [
"7",
"2",
"9",
"6"
] | 6 | attribute type: color
attribute value: orange
count: 6
grid number: 3
question type: how many
target category: orange | |
22-2 | How many orange orange are there in the image? | [
"2",
"6",
"1",
"4"
] | 6 | attribute type: color
attribute value: orange
count: 6
grid number: 3
question type: how many
target category: orange | |
23-0 | How many orange are there in the image? | [
"7",
"2",
"4",
"6"
] | 7 | count: 7
grid number: 3
question type: how many
target category: orange | |
23-1 | How many orange are there in the image? | [
"6",
"9",
"8",
"7"
] | 7 | count: 7
grid number: 3
question type: how many
target category: orange | |
23-2 | How many orange are there in the image? | [
"6",
"8",
"1",
"7"
] | 7 | count: 7
grid number: 3
question type: how many
target category: orange | |
24-0 | How many orange objects are there in the image? | [
"1",
"7",
"4",
"8"
] | 4 | attribute type: color
attribute value: orange
count: 4
grid number: 3
question type: how many | |
24-1 | How many orange objects are there in the image? | [
"7",
"8",
"3",
"4"
] | 4 | attribute type: color
attribute value: orange
count: 4
grid number: 3
question type: how many | |
24-2 | How many orange objects are there in the image? | [
"1",
"4",
"9",
"5"
] | 4 | attribute type: color
attribute value: orange
count: 4
grid number: 3
question type: how many | |
25-0 | How many metal sugary drink are there in the image? | [
"6",
"2",
"4",
"1"
] | 6 | attribute type: material
attribute value: metal
count: 6
grid number: 3
question type: how many
target category: sugary drink | |
25-1 | How many metal sugary drink are there in the image? | [
"5",
"1",
"6",
"8"
] | 6 | attribute type: material
attribute value: metal
count: 6
grid number: 3
question type: how many
target category: sugary drink | |
25-2 | How many metal sugary drink are there in the image? | [
"1",
"4",
"6",
"3"
] | 6 | attribute type: material
attribute value: metal
count: 6
grid number: 3
question type: how many
target category: sugary drink | |
26-0 | How many beige container are there in the image? | [
"3",
"1",
"4",
"2"
] | 4 | attribute type: color
attribute value: beige
count: 4
grid number: 2
question type: how many
target category: container | |
26-1 | How many beige container are there in the image? | [
"3",
"2",
"4",
"1"
] | 4 | attribute type: color
attribute value: beige
count: 4
grid number: 2
question type: how many
target category: container | |
26-2 | How many beige container are there in the image? | [
"1",
"4",
"3",
"2"
] | 4 | attribute type: color
attribute value: beige
count: 4
grid number: 2
question type: how many
target category: container | |
27-0 | How many beige tool are there in the image? | [
"9",
"2",
"5",
"8"
] | 9 | attribute type: color
attribute value: beige
count: 9
grid number: 3
question type: how many
target category: tool | |
27-1 | How many beige tool are there in the image? | [
"8",
"2",
"9",
"6"
] | 9 | attribute type: color
attribute value: beige
count: 9
grid number: 3
question type: how many
target category: tool | |
27-2 | How many beige tool are there in the image? | [
"7",
"9",
"4",
"1"
] | 9 | attribute type: color
attribute value: beige
count: 9
grid number: 3
question type: how many
target category: tool | |
28-0 | How many orange food are there in the image? | [
"3",
"7",
"4",
"2"
] | 4 | attribute type: color
attribute value: orange
count: 4
grid number: 3
question type: how many
target category: food | |
28-1 | How many orange food are there in the image? | [
"8",
"1",
"4",
"2"
] | 4 | attribute type: color
attribute value: orange
count: 4
grid number: 3
question type: how many
target category: food | |
28-2 | How many orange food are there in the image? | [
"5",
"4",
"2",
"6"
] | 4 | attribute type: color
attribute value: orange
count: 4
grid number: 3
question type: how many
target category: food | |
29-0 | How many white egg are there in the image? | [
"2",
"1",
"4",
"6"
] | 6 | attribute type: color
attribute value: white
count: 6
grid number: 3
question type: how many
target category: egg | |
29-1 | How many white egg are there in the image? | [
"6",
"4",
"1",
"8"
] | 6 | attribute type: color
attribute value: white
count: 6
grid number: 3
question type: how many
target category: egg | |
29-2 | How many white egg are there in the image? | [
"7",
"6",
"8",
"3"
] | 6 | attribute type: color
attribute value: white
count: 6
grid number: 3
question type: how many
target category: egg | |
30-0 | How many wood step stool are there in the image? | [
"9",
"3",
"5",
"8"
] | 9 | attribute type: material
attribute value: wood
count: 9
grid number: 3
question type: how many
target category: step stool | |
30-1 | How many wood step stool are there in the image? | [
"7",
"9",
"4",
"5"
] | 9 | attribute type: material
attribute value: wood
count: 9
grid number: 3
question type: how many
target category: step stool | |
30-2 | How many wood step stool are there in the image? | [
"7",
"9",
"8",
"1"
] | 9 | attribute type: material
attribute value: wood
count: 9
grid number: 3
question type: how many
target category: step stool | |
31-0 | How many black container are there in the image? | [
"2",
"7",
"3",
"1"
] | 7 | attribute type: color
attribute value: black
count: 7
grid number: 3
question type: how many
target category: container | |
31-1 | How many black container are there in the image? | [
"6",
"2",
"7",
"1"
] | 7 | attribute type: color
attribute value: black
count: 7
grid number: 3
question type: how many
target category: container | |
31-2 | How many black container are there in the image? | [
"9",
"4",
"3",
"7"
] | 7 | attribute type: color
attribute value: black
count: 7
grid number: 3
question type: how many
target category: container | |
32-0 | How many gray container are there in the image? | [
"6",
"1",
"5",
"9"
] | 6 | attribute type: color
attribute value: gray
count: 6
grid number: 3
question type: how many
target category: container | |
32-1 | How many gray container are there in the image? | [
"3",
"6",
"2",
"4"
] | 6 | attribute type: color
attribute value: gray
count: 6
grid number: 3
question type: how many
target category: container | |
32-2 | How many gray container are there in the image? | [
"1",
"2",
"6",
"9"
] | 6 | attribute type: color
attribute value: gray
count: 6
grid number: 3
question type: how many
target category: container | |
33-0 | How many brown crate are there in the image? | [
"6",
"2",
"8",
"9"
] | 8 | attribute type: color
attribute value: brown
count: 8
grid number: 3
question type: how many
target category: crate |
End of preview. Expand
in Dataset Viewer.
Dataset Card for TaskMeAnything-v1-imageqa-2024
TaskMeAnything-v1-imageqa-2024 benchmark dataset
🌐 Website | 📑 Paper | 🤗 Huggingface | 💻 Interface
If you like our project, please give us a star ⭐ on GitHub for latest update.
TaskMeAnything-v1-2024
TaskMeAnything-v1-imageqa-2024 is a benchmark for reflecting the current progress of MLMs by automatically finding tasks that SOTA MLMs struggle with using the TaskMeAnything Top-K queries.
This benchmark includes 3,279 2d questions, 7,095 3d questions, and 1,896 real image questions that the TaskMeAnything algorithm automatically approximated as challenging for over 12 popular MLMs.
The dataset contains 19 splits, while each splits contains 600+ questions from a specific task generator in TaskMeAnything-v1. For each row of dataset, it includes: image, question, options, answer and its corresponding task plan.
Load TaskMeAnything-v1-2024 ImageQA Dataset
import datasets
dataset_name = 'weikaih/TaskMeAnything-v1-imageqa-2024'
dataset = datasets.load_dataset(dataset_name, split = TASK_GENERATOR_SPLIT)
where TASK_GENERATOR_SPLIT is one of the task generators, eg, 2024_2d_how_many.
Evaluation Results
Overall
Breakdown performance on each task types
Out-of-Scope Use
This dataset should not be used for training models.
Disclaimers
TaskMeAnything and its associated resources are provided for research and educational purposes only. The authors and contributors make no warranties regarding the accuracy or reliability of the data and software. Users are responsible for ensuring their use complies with applicable laws and regulations. The project is not liable for any damages or losses resulting from the use of these resources.
Contact
- Jieyu Zhang: jieyuz2@cs.washington.edu
Citation
BibTeX:
@article{zhang2024task,
title={Task Me Anything},
author={Zhang, Jieyu and Huang, Weikai and Ma, Zixian and Michel, Oscar and He, Dong and Gupta, Tanmay and Ma, Wei-Chiu and Farhadi, Ali and Kembhavi, Aniruddha and Krishna, Ranjay},
journal={arXiv preprint arXiv:2406.11775},
year={2024}
}
- Downloads last month
- 44