Fresh Food Product Specification and Variant Comparison QA Dataset

#question answering #product attribute comparison #product variant identification #fresh food e-commerce #grocery product comparison #food specification and variant selection

5000
0.4GB
JSON
CATL
MOBIUSI INC

Updated:2026-10-26

AI Analysis & Value Prop

This dataset provides retrieval and comparison questions about same-brand or same-series products in grocery and fresh food delivery catalogs, covering differences in size, flavor, and packaging. Each example links relevant catalog records with an extracted comparison target, attribute-level findings, an answer, and supporting evidence, helping models identify the correct variant and respond from catalog information. It is suited to training and evaluating retrieval-augmented generation, product catalog question answering, and shopping assistants.

Dataset Insights

Sample Examples

{}

Technical Specifications

Field	Type	Description
file_name	string	File name
answer	string	The final response, grounded in the relevant catalog records and comparison findings.
evidence	array	Excerpts from catalog records that jointly support the answer.
question	string	The original question requesting retrieval, identification, or comparison of product variants.
catalog_records	array	The original product catalog listings available as evidence for answering the question.
comparison_target	object	The product identity and variant criteria extracted from the question to guide catalog retrieval and comparison.
comparison_findings	array	Attribute-by-attribute comparison findings derived from the selected catalog listings.
selected_catalog_codes	array	References identifying the catalog records selected as relevant to the question.

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 dataset contain?: It covers variants of same-brand or same-series grocery and fresh food products, including differences in size, flavor, and packaging. Examples include source catalog records, user questions, comparison targets, selected catalog codes, comparison findings, answers, and supporting evidence.

How can it support product question-answering?: Examples require retrieving and comparing information across relevant catalog listings, then answering from evidence that supports the response. They can train or evaluate product variant retrieval, attribute comparison, and evidence-grounded answering.

How can I access the data, and what are the licensing terms?: Check the dataset product page or the provider's channel for access instructions, available formats, and license terms. Confirm permitted use before working with the data, including any requirements for commercial use, redistribution, or privacy.

Which applications is it suitable for?: Suitable applications include product search and shopping assistants for grocery delivery platforms, catalog question answering, distinguishing similar products, and training or evaluating retrieval-augmented generation systems.

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={Fresh Food Product Specification and Variant Comparison QA Dataset},
  author={MOBIUSI INC},
  year={2026},
  url={https://www.mobiusi.com/datasets/fddb8fd48256426d430bd55af8dadde3},
  urldate={2026-10-20},
  keywords={fresh food QA dataset,product specification comparison,product variant identification,grocery delivery catalog,catalog retrieval QA,RAG training data,product attribute comparison,evidence-grounded QA},
  version={1.0}
}

Using this in research? Please cite us.

placeholder

Products

Scene

Domain

Modality

Task