Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,48 @@
|
|
1 |
-
---
|
2 |
-
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
license: cc-by-nc-nd-4.0
|
5 |
+
configs: # Optional. This can be used to pass additional parameters to the dataset loader, such as `data_files`, `data_dir`, and any builder-specific parameters
|
6 |
+
- config_name: default # Name of the dataset subset, if applicable. Example: default
|
7 |
+
data_files:
|
8 |
+
- split: train # Example: train
|
9 |
+
path: train/* # Example: data.csv
|
10 |
+
- split: dev # Example: test
|
11 |
+
path: dev/* # Example: holdout.csv
|
12 |
+
- split: test # Example: test
|
13 |
+
path: test/* # Example: holdout.csv
|
14 |
+
dataset_info:
|
15 |
+
features:
|
16 |
+
- name: video_path # Example: id
|
17 |
+
dtype: string # Example: int32
|
18 |
+
- name: abstract # Example: text
|
19 |
+
dtype: string # Example: string
|
20 |
+
- name: language # Example: image
|
21 |
+
dtype: string # Example: image
|
22 |
+
- name: split # Example: image
|
23 |
+
dtype: string # Example: image
|
24 |
+
- name: conference # Example: image
|
25 |
+
dtype: string # Example: image
|
26 |
+
- name: year # Example: image
|
27 |
+
dtype: string # Example: image
|
28 |
+
config_name: default # Name of the dataset subset. Example for glue: sst2
|
29 |
+
splits:
|
30 |
+
- name: train # Example: train
|
31 |
+
num_examples: 4333 # Example for SQuAD: 87599
|
32 |
+
- name: dev # Example: train
|
33 |
+
num_examples: 886 # Example for SQuAD: 87599
|
34 |
+
- name: test # Example: train
|
35 |
+
num_examples: 1432 # Example for SQuAD: 87599
|
36 |
+
---
|
37 |
+
|
38 |
+
# Abstract Generation with ACL talks
|
39 |
+
|
40 |
+
This dataset is designed for the task of Multimodal Abstract Generation from scientific talks.
|
41 |
+
It provides links to videos of academic presentations from ACL conferences, paired with their corresponding abstracts.
|
42 |
+
The goal is to generate concise, informative summaries (abstracts) based on the content.
|
43 |
+
|
44 |
+
Soon, we will also upload the audio files for easier usage!
|
45 |
+
|
46 |
+
## Citation
|
47 |
+
|
48 |
+
A dataset paper will be available soon.
|