Agentic data science with the Data Agent Kit
cloud.google.com·6h ago
AI Research
arXiv CS.AI·Aug 20·5 min read
Language Models for Portuguese: A Systematic Mapping Study
TL;DR
A systematic mapping study identifies and characterizes 46 language models developed for Portuguese, highlighting their diverse architectures and training resources. This comprehensive overview reveals research gaps and future opportunities in the field.
✦ Why It Matters
Explore the identified research gaps to guide your next project in Portuguese language modeling.
Key Takeaways
- Review the 46 mapped language models for insights.
- Identify gaps in current Portuguese NLP research.
- Consider collaboration to enhance model development.
- Utilize the comprehensive overview for project planning.
Full Summary
Language models have significantly advanced Natural Language Processing (NLP), but their development varies by language. In Portuguese, recent academic and corporate efforts have led to a growing ecosystem of language models.
This study systematically maps 46 distinct models, analyzing aspects such as base model, architecture, computational resources, training datasets, licensing, and code availability. By employing a phylogenetic perspective, the research illustrates the evolution and interrelationships among these models.
Key findings include the identification of research gaps and opportunities for further development, emphasizing the need for more robust datasets and collaborative efforts. The study serves as a foundational resource for researchers and developers in the Portuguese NLP community.