> ## Documentation Index
> Fetch the complete documentation index at: https://docs.cortex-labs.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Understand mentions, entities, and relationships

> Understand source statements, workspace interpretation and the limits of counts, confidence and time views.

The world model connects source statements with workspace lists. It helps you inspect related material while keeping evidence and interpretation distinct.

## A mention records a source statement

Extraction produces mentions from versioned fragments. Each mention identifies its enrichment result and exact source fragment version.

| Mention | Records |
| - | - |
| Entity | A named or identifiable thing, its type and its source locator. |
| Relationship | A connection between entity mentions in the same result, with a locator. |
| Attribute | A stated value, unit and locator for an entity or relationship. |

A mention belongs to a dataset. Workspaces reaching that dataset can use the same eligible mentions. Source access still applies.

You cannot edit a mention. Correct the workspace's interpretation through curation, or correct and reprocess the source.

## Lists express the workspace's current interpretation

Each workspace maintains canonical entities and relationships. Aliases connect mentions to these entries. Two similar names need not identify one entity.

A list entry can be **New**, **Accepted** or **Hidden**. People can also add entries directly. Acceptance records a curation decision; it does not prove the underlying claim.

The workspace pins a domain-schema version. That schema supplies layers, types, attributes, units and permitted relationship endpoints. Unlisted types appear under **Other**.

Publishing another schema does not move the workspace's pin. An Owner must select the new version deliberately.

## Evidence counts describe coverage

| Count | What it counts |
| - | - |
| Mention | Extracted statements the reader can access. |
| Observation | Distinct root source items, such as posts or a video. |
| Source | Producing ingestion jobs, or the upload dataset when no job identifies the source. |

Several analyses of one item do not create independent observations. Several frames of one video still count as one observation. Counts do not establish independent corroboration.

Competing attribute values remain separate. A confidence score describes its stated method; it is not automatically a calibrated probability.

## Time views answer different questions

Event time describes when a statement applies. Collected time describes when the platform first saved its fragment. A re-scrape keeps that first collection time.

The lists show their current interpretation under the chosen time filter. They do not reconstruct every earlier state of curation. Use the edit log to inspect decisions and [report-edition comparison](/vantage/review-and-publish/compare-editions) to compare published assessments.

Start with [Curate world-model lists and resolve identities](/vantage/explore-evidence/curate-identities) or [Follow a finding to its evidence](/vantage/explore-evidence/follow-evidence).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.