Cross-Merchant Beauty Service Catalog Evidence Question Answering Benchmark

#evidence-based question answering #service information retrieval #merchant identification #beauty service discovery #cross-merchant service comparison #intelligent beauty service assistants

5000
0.3GB
JSON
CATL
MOBIUSI INC

Updated:2026-10-19

AI Analysis & Value Prop

This benchmark contains original catalog text from multiple beauty and hair service merchants, questions about service offerings and merchant identity, and corresponding answers with evidence annotations. It records the target merchant, relevant service, catalog evidence quote, evidence merchant, merchant-match status, and whether the answer is supported, enabling evaluation of retrieval and evidence attribution in multi-merchant catalog settings. Typical applications include beauty service discovery, catalog question answering, merchant entity disambiguation, and answer traceability analysis.

Dataset Insights

Sample Examples

{}

Technical Specifications

Field	Type	Description
file_name	string	File name
answer	string	An answer based on the catalog information; if the catalog cannot support an answer, this field should clearly state that it cannot be determined.
service_name	string	The service name identified from the question or catalog that is relevant to the answer.
user_question	string	A question to answer using the merchant catalog, potentially concerning a service offering or merchant distinction.
evidence_quote	string	An excerpt from the catalog that directly supports the answer; leave blank when the catalog contains no relevant evidence.
target_merchant	string	The beauty or hair service merchant identified as the target of the question.
catalog_document	string	Original catalog text containing beauty and hair service merchants and their offerings; it may include information from multiple merchants.
evidence_merchant	string	The merchant associated with the evidence excerpt, used to verify whether the correct merchant's catalog information was cited.
is_merchant_match	boolean	Indicates whether the merchant associated with the evidence matches the target merchant in the question.
is_answer_supported	boolean	Indicates whether the cited catalog evidence directly supports the answer.

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 the benchmark contain?: Each record includes original merchant catalog text, a user question, the target merchant, a relevant service name, and an answer. It also provides the supporting catalog excerpt and its merchant, along with labels for merchant matching and answer support, making retrieval and attribution checks possible.

How can answer reliability be evaluated?: Compare each answer with its catalog evidence quote and verify that the evidence merchant matches the target merchant in the question. When the catalog does not support an answer, support-status and evidence fields make unsupported responses and cross-merchant confusion easier to assess.

Which applications is this data suited for?: The benchmark is suited to beauty service discovery, merchant catalog search, retrieval-augmented question answering evaluation, knowledge grounding, and merchant entity disambiguation. It can also be used to compare evidence citation accuracy and answer traceability.

How can the data be obtained, and where can licensing terms be checked?: Check the dataset product page for available access methods, data formats, and licensing terms. Commercial use, redistribution, and other uses are governed by the license published on that page, so review its restrictions before use.

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={Cross-Merchant Beauty Service Catalog Evidence Question Answering Benchmark},
  author={MOBIUSI INC},
  year={2026},
  url={https://www.mobiusi.com/datasets/3c1ae564441b64519524bf7db3e8aa02},
  urldate={2026-10-10},
  keywords={beauty service catalogs, beauty service QA, multi-merchant retrieval, evidence QA benchmark, RAG evaluation, knowledge grounding, merchant disambiguation, catalog question answering},
  version={1.0}
}

Using this in research? Please cite us.

placeholder

Products

Scene

Domain

Modality

Task