G

Gliner Medium News V2.1

Developed by EmergentMethods
A fine-tuned version based on GLiNER, optimized for news entity extraction, achieving up to 7.5% higher zero-shot accuracy across 18 benchmark tests
Downloads 532.81k
Release Time : 4/17/2024

Model Overview

An entity recognition model optimized for long-text news scenarios, supporting cross-domain topic identification, with synthetic datasets ensuring global diverse perspectives

Model Features

Cross-domain Performance Improvement
Achieves up to 7.5% higher zero-shot accuracy than the base model across 18 benchmark datasets
Global Diverse Data
Production Environment Friendly
Compact size suitable for high-throughput scenarios, already adopted by the AskNews system

Model Capabilities

News Entity Recognition
Multilingual Text Processing
Long-text Entity Extraction
Zero-shot Transfer Learning

Use Cases

News Analysis
International News Event Analysis
Extract key entities such as people, locations, and organizations from multilingual news
Accurately identifies entity relationships across cultural contexts
Crime Report Parsing
Extract suspect names, times, locations, tools, and other case-related information
Successfully identifies complex entities like vehicle models, ages, and institution names in examples
Content Moderation
Sensitive Entity Screening
Automatically detect sensitive person or organization names in news
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase