---
license: apache-2.0
task_categories:
- question-answering
---
# A Controllable Examination for Long-Context Language Models

This is a sample dataset created using Qwen-7B tokenizer. For details, please check our [paper](https://arxiv.org/abs/2506.02921)

This dataset is created by running from our [repo](https://github.com/Thomasyyj/LongBio-Benchmark/tree/main)

```
bash scripts/submit_scripts/prepare_data.sh
```