Dataset · Neurips-2026-MLIntern-Repro
NeurIPS 2026 scientific papers
Neurips-2026-MLIntern-Repro/papers
English dataset of NeurIPS 2026 papers with key claims for reproduction.
What's inside
Contains 1046 papers with metadata (arXiv ID, authors, presentation type, arXiv and OpenReview links). Each record includes one key claim that an agent can attempt to reproduce or refute. The size is between 1 thousand and 10 thousand rows.
What you can do with it
- Developing and testing scientific AI agents for research automation
- Benchmarking text generation and fact verification systems
- Tasks
- Text generation
- Languages
- English
- Size
- 1K–10K rows
- Downloads this month
- 1.7K
- Likes
- 2
How to load it
from datasets import load_dataset
ds = load_dataset("Neurips-2026-MLIntern-Repro/papers")