Dataset · FiscalNote
US bill corpus with short summaries
FiscalNote/billsum
10k to 100k texts of US Congress and California state bills with annotations for summarization tasks.
What's inside
English text dataset containing 10k to 100k examples of US Congress and California state bills. Each record includes the full document text, its summary, as well as text and annotation length in characters; titles are additionally specified for federal laws.
What you can do with it
- Training or fine-tuning (LoRA) models for summarizing long legal texts
- Testing compression algorithms for official and legislative documents
- Tasks
- Summarization
- Languages
- English
- Size
- 10K–100K rows
- Downloads this month
- 8K
- Likes
- 57
Models trained on it
- luluw/t5-base-finetuned-billsum 29
- Anurag33Gaikwad/legal-led-billsum-summarizationSummarization 26
- Driw0x/my_awesome_billsum_modelSummarization 22
- getallineedeasily/billsum-model-z 21
- Anurag33Gaikwad/legal-pegasus-billsum-summarizationSummarization 21
- solvi-ai/solvi-large-longText classification 15
- aun09/flan-t5-legal-summary 14
- adithya-06/my_awesome_billsum_model 5
How to load it
from datasets import load_dataset
ds = load_dataset("FiscalNote/billsum")