LightTag

Upload PDF/Text contracts.

Define entity schema (Parties, Dates, Clauses).

Distribute to legal paralegals.

Use Relationship extraction to link parties to specific obligations.

Export to JSON for training a contract AI.

Import clinical notes.

Define PII entities (Names, SSN, Locations).

Enable dual-consensus to ensure 100% catch rate.

Review discrepancies between annotators.

Export scrubbed text.

Stream RSS/News feeds into LightTag.

Label entities (Tickers, Indicators) and sentiment (Bullish/Bearish).

Use Active Learning to surface high-impact news.

Train custom BERT model on the output.

Export tickets from Zendesk.

Import to LightTag.

Label Intent and Sub-intent.

Analyze labeler agreement to refine category definitions.

Train NLU classifier.

Upload product descriptions in multiple languages.

Assign native speakers to specific language jobs.

Use the 'Dictionary' feature for standard brand names.

Export unified JSON for cross-lingual model training.

Define complex nested entities (Chemical -> Element).

Link compounds to effects via Relationship labeling.

Validate using subject matter experts (SMEs).

Generate a knowledge graph from the exported data.

Upload conversation logs.

Label 'Slot filling' and 'Turn intents'.

Use hierarchical labels for complex dialogue paths.

Review flow consistency using the 'Conflict resolution' view.