Extract cleanroom classifications, environmental conditions, services, and equipment from facility webpages to support resource discovery and catalog building.

Evaluate how well models extract key accuracy and calibration claims from specialized temperature sensor web pages.

Turn road signs, lane restrictions, vehicle classes, and time conditions into queryable rule relations for Robotaxi maps.

Convert component descriptions and fabrication requirements into structured specifications for prefabricated construction data management and model training.

Turn luxury hotel room and suite descriptions into structured attributes for accommodation catalog extraction and model training.

Extract text-grounded attributes from menu names and descriptions to support menu structuring and model fine-tuning.

Turn contact center routing instructions into structured constraints to help automate customer service routing configuration.

Extract component-to-material relationships from footwear product descriptions with source evidence for traceable, structured catalog data.

The Pharmacy Prescription Text Extraction Dataset focuses on extracting and recognizing text data from medical prescription images.

Explore the key contract mechanisms for supply disruptions and legal changes, with structured insights into risk allocation and performance consequences.

Extract earnings highlights from financial news and link every structured result to precise source evidence.

Connect process events with orders, invoices, and deliveries to build knowledge graphs that represent cross-object process relations.

Extract operating conditions, contact states, and related sensor-state information from temperature sensor web text for focused domain evaluation.

Transform virtual lab safety notices and procedure requirements into structured rules for accurate safety constraint extraction.

Extract service and resource entities from APM catalogs, inventories, and telemetry records to build reliable structured-output capabilities.

Evaluate how well models extract and structure key flow sensor specifications from product webpages and technical materials, with field-level annotations and linked source text.

Evaluate how well models extract maturity and harvest information from banana production text, including harvest conditions and postharvest measures.

Train models to identify Korean vocabulary information and infer contextual senses using evidence from the source text.

Turn experiment protocols into structured information about reagents, equipment, steps, and conditions for virtual laboratory workflows.

Train models to understand local activity booking requests and generate structured availability-query arguments.