Bible Cross-Chapter Cross-Reference Multi-Hop QA Dataset

#Multi-Hop Question Answering #Cross-Chapter Reasoning #Scriptural Evidence Citation #Reasoning Step Generation #Religious Text QA #Scripture Retrieval and Linking #Biblical Text Research #Religious Education

3610
0.3GB
JSON
CATL
MOBIUSI INC

Updated:2026-11-04

AI Analysis & Value Prop

This dataset is built from Bible passages with explicitly identified editions and translations, creating multi-hop questions around cross-chapter references, quotations, and parallel passages. Each record includes a question, answer, reasoning steps, textual evidence, cross-reference links, and a hop count to make the answer path traceable. It supports training and evaluating models on biblical text relationships, evidence-based reasoning, and multi-hop question answering.

Dataset Insights

Sample Examples

{}

Technical Specifications

Field	Type	Description
file_name	string	File name
answer	string	A concise answer supported by the supplied Bible passages and their internal relationships.
question	string	A question that requires combining information from multiple supplied passages or their explicit relationships.
hop_count	integer	The number of distinct inferential transitions needed to connect the source evidence to the answer.
bible_edition	string	The explicitly identified Bible translation or edition used for the source passages.
reasoning_steps	array	An ordered explanation of how the answer follows from the source passages and cross-reference links.
source_passages	array	The Bible passages provided as source material for answering the question.
evidence_citations	array	Passage-level textual evidence that directly supports the answer or key reasoning steps.
cross_reference_links	array	Explicit links among the supplied passages, including quotations, citations, or parallel passages.

Compliance Statement

Authorization Type	Proprietary - Commercial AI Training License (No Redistribution)
Commercial Use	Requires exclusive subscription or authorization contract (monthly or per-invocation charging)
Privacy and Anonymization	No PII, no real company names, simulated scenarios follow industry standards
Compliance System	Compliant with China's Data Security Law / EU GDPR / supports enterprise data access logs

Frequently Asked Questions

What does each record contain?: Each record includes the Bible edition, source passages, cross-reference links, a question, an answer, reasoning steps, evidence citations, and a hop count. The data shows how answers follow from multiple passages and their relationships.

How can users verify the answers?: Records provide passage-level citations, explicit cross-reference links, and ordered reasoning steps. These details let users check whether an answer is supported by the cited text and whether the reasoning follows the recorded relationships.

Which tasks is the dataset suited for?: It is suited for multi-hop Bible question answering, cross-chapter reasoning, scriptural evidence citation, reasoning trace generation, and training or evaluation of religious text QA models.

How can the dataset be obtained, and what licensing should users consider?: Refer to the product page for the available delivery method, file specifications, and dataset license. Because records use specific Bible editions or translations, users should also verify the relevant text rights and comply with the applicable terms.

Can't find the data you need?

Post a request and let data providers reach out to you.

Get this Dataset

Verified for Enterprise Use

Cite this Work

@dataset{Mobiusi2026,
  title={Bible Cross-Chapter Cross-Reference Multi-Hop QA Dataset},
  author={MOBIUSI INC},
  year={2026},
  url={https://www.mobiusi.com/datasets/a89fe4fc9edac68948346fa034d283c2},
  urldate={2026-10-26},
  keywords={Bible QA dataset,Bible cross-reference data,multi-hop question answering,scripture reasoning,Bible passage linking,religious text dataset,reasoning SFT data,evidence-based QA},
  version={1.0}
}

Using this in research? Please cite us.

placeholder

Scene

Domain

Modality

Task