Virtual Laboratory Experiment State Fact Verification Benchmark Dataset
- Records
- 5600
- Size
- 0.3GB
- Format
- JSON
- Provider
MOBIUSI INC- Updated
- v1.02026-11-04
This JSON dataset records experiment states and changes in virtual laboratories, including initial conditions, sequences of interactive actions, state transitions, final states, and natural-language or structured fact claims. It supports training and evaluating models on whether they can verify facts such as whether a reagent was added, equipment is running, or a sample meets a specified condition, while grounding conclusions in trajectory evidence. It is suitable for developing and evaluating virtual lab agents, state-tracking systems, and factual reasoning models.
Overview
Dataset Insights
Preview
Sample Examples
Schema
Technical Specifications
| Field | Type | Description |
|---|---|---|
| file_name | string | File name |
| trajectory_summary | object | Summary statistics for the experiment trajectory, including action counts, claim verification counts, and final consistency. |
| experiment_trajectory | object | Raw JSON trajectory from a virtual laboratory experiment, including the initial state, action sequence, state changes, final state, and claims. |
| fact_verification_results | array | Verification outcomes and supporting evidence calculated for each factual claim based on the experiment trajectory. |
Trust
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
FAQ
Frequently Asked Questions
- What experimental information does the dataset contain?
- Each experiment trajectory may include initial conditions, sequences of interactive actions, state changes, a final state, and natural-language or structured fact claims. The dataset also provides claim verification results, supporting evidence, and trajectory summaries.
- Which capabilities can this dataset evaluate?
- It can evaluate whether a model tracks changes throughout an experiment, verifies specific factual claims, and identifies relevant evidence. It can also support training for fact verification and state reasoning systems.
- How can I obtain the dataset?
- Use the download or request option provided on this product page, and refer to the listed delivery format and usage instructions.
- What are the licensing terms and permitted uses?
- The applicable license and usage scope are defined by the product page or the license provided with the data. Confirm the relevant terms before using the dataset for research, model training, or commercial projects.
Can't find the data you need?
Post a request and let data providers reach out to you.
Get this Dataset
High-quality data, ready for your next AI breakthrough.
Cite this Work
@dataset{Mobiusi2026,
title={Virtual Laboratory Experiment State Fact Verification Benchmark Dataset},
author={MOBIUSI INC},
year={2026},
url={https://www.mobiusi.com/datasets/1f117f8e0b3f9cda03646d365570f179?dataset_scene_id=7},
urldate={2026-10-27},
keywords={virtual laboratory dataset,experiment state verification,fact verification benchmark,experiment trajectory data,lab agent evaluation,state transition reasoning,JSON benchmark data,claim verification},
version={1.0}
}Using this in research? Please cite us.