ismailemir/arxiv-corpus

ArXiv Paper Abstracts

Collection of ArXiv paper abstracts in JSON format for search and retrieval tasks.

📊 Dataset Info

• Papers: 2+ million
• Format: JSON Lines (one object per line)
• Source: Kaggle ArXiv Dataset
• Updated: Version 266

📝 Format

Each line contains a JSON object:

{"abstract": "Paper abstract text..."}

🚀 Quick Start

from huggingface_hub import hf_hub_download
import json

# Download corpus
corpus_path = hf_hub_download(
    repo_id="ismailemir/arxiv-corpus",
    filename="arxiv_abstracts.json",
    repo_type="dataset"
)

# Load corpus
corpus = []
with open(corpus_path, 'r') as f:
    for line in f:
        row = json.loads(line)
        corpus.append(row["abstract"])

print(f"Loaded {len(corpus):,} abstracts")

💡 Use Cases

• 📚 Academic paper search
• 🔍 Information retrieval research
• 🤖 Training search models
• 📊 Text mining and analysis
• 🧪 Benchmarking retrieval systems

🔗 Related

• 🔍 Search Indices: ismailemir/arxiv-indices
• 🔬 Original Dataset: Cornell ArXiv

📄 License

Apache 2.0 - Please cite ArXiv if using this data.

🙏 Citation

@article{clement2019arxiv,
  title={On the Use of ArXiv as a Dataset},
  author={Clement, Colin B and Bierbaum, Matthew and O'Keeffe, Kevin P and Alemi, Alexander A},
  journal={arXiv preprint arXiv:1905.00075},
  year={2019}
}

📧 Contact

For issues or questions, please open an issue on the Hugging Face dataset page.