Dataset Viewer
Auto-converted to Parquet Duplicate
document_id
string
document_text
string
document_filename
string
document_metadata
dict
document_summary
string
summarization_model
string
chunks
list
multihop_chunks
list
b7a2fa52-f136-44da-bf66-02196abe4c4b
"5\n2\n0\n2\n\nr\np\nA\n2\n\n]\nL\nC\n.\ns\nc\n[\n\n1\nv\n3\n3\n8\n1\n0\n.\n4\n0\n5\n2\n:\nv\ni\nX\n(...TRUNCATED)
yourbench_arxiv_paper.md
{ "file_size": 133539 }
"YourBench is an open-source, automated framework that generates dynamic, domain-specific, and relia(...TRUNCATED)
Qwen/Qwen3-4B-Instruct-2507
[{"chunk_id":"b7a2fa52-f136-44da-bf66-02196abe4c4b_0","chunk_text":"5\n2\n0\n2\n\nr\np\nA\n2\n\n]\nL(...TRUNCATED)
[{"chunk_ids":["b7a2fa52-f136-44da-bf66-02196abe4c4b_63","b7a2fa52-f136-44da-bf66-02196abe4c4b_88"],(...TRUNCATED)

Built with YourBench

Test Extensive Custom Schema Demo

This dataset was generated using YourBench (v0.6.0), an open-source framework for generating domain-specific benchmarks from document collections.

Pipeline Steps

  • ingestion: Read raw source documents, convert them to normalized markdown and save for downstream steps
  • chunking: Split texts into token-based single-hop and multi-hop chunks
  • single_shot_question_generation: Generate standalone question-answer pairs per chunk using LLM

Reproducibility

To reproduce this dataset, use YourBench v0.6.0 with the following configuration:

hf_configuration:
  hf_dataset_name: test_extensive_custom_schema_demo
  hf_organization: $HF_ORGANISATION
  push_to_hub: true
model_list:
- model_name: Qwen/Qwen3-4B-Instruct-2507
  base_url: http://us-east-1.sumuk.org:9100/stargate.neon-pentatonic.ts.net/30000/v1/
  api_key: $API_KEY
  max_concurrent_requests: 8
pipeline:
  ingestion:
    source_documents_dir: test_extensive_schema
    output_dir: data/processed
    supported_file_extensions:
    - .md
    - .txt
    - .pdf
  chunking:
    l_max_tokens: 300
    token_overlap: 100
  single_shot_question_generation:
    single_shot_system_prompt: custom_single_shot_system_prompt.md
    single_shot_user_prompt: custom_single_shot_user_prompt.md
    chunk_sampling:
      enable: false
      num_samples: 100
      strategy: random
      random_seed: 42
    use_structured_outputs: false
    structured_fallback: true
    custom_schema_path: test_extensive_schema/schema.py
    custom_schema_class: ExtensiveQuestion
    custom_schema_auto_batch: true

(This dataset card was automatically generated by YourBench)

Downloads last month
5