Datasets:
Tasks:
Summarization
Modalities:
Text
Formats:
json
Sub-tasks:
text2text-generation
Languages:
English
Size:
10K - 100K
ArXiv:
License:
Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -46,6 +46,14 @@ This format is commonly referred to as:
|
|
| 46 |
- `AIC`
|
| 47 |
- `FullText`
|
| 48 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 49 |
## Preparation summary
|
| 50 |
|
| 51 |
- Task: one-sentence scientific TLDR generation.
|
|
|
|
| 46 |
- `AIC`
|
| 47 |
- `FullText`
|
| 48 |
|
| 49 |
+
## Original Dataset Highlights
|
| 50 |
+
|
| 51 |
+
- Original dataset: `allenai/scitldr`
|
| 52 |
+
- Focus: extreme summarization of scientific papers (TLDR generation).
|
| 53 |
+
- Reported scale on source card: 5.4K TLDRs over ~3.2K papers.
|
| 54 |
+
- Multi-target setup: each paper can have multiple valid TLDR summaries.
|
| 55 |
+
- Paper: [TLDR: Extreme Summarization of Scientific Documents](https://arxiv.org/abs/2004.15011)
|
| 56 |
+
|
| 57 |
## Preparation summary
|
| 58 |
|
| 59 |
- Task: one-sentence scientific TLDR generation.
|