Third-party cyber evaluations involving OpenAI models
openai.com·14h ago
AI Research
arXiv CS.AI·Jun 9·5 min read
AgriGov: A Structured Multilingual Dataset Curation for Indian Government Schemes for Farmers
TL;DR
A lack of multilingual resources for Indian agricultural policies hindered effective communication with farmers. AgriGov was developed as a trilingual dataset (English-Hindi-Marathi) by curating data from 50 government schemes and enhancing it with additional sentence pairs.
✦ Why It Matters
Engineers can leverage AgriGov to build better multilingual applications for agricultural information dissemination.
Key Takeaways
- Leverage AgriGov for multilingual machine translation projects.
- Utilize structured data for developing farmer-facing applications.
- Incorporate human post-editing for improved translation accuracy.
- Explore the dataset for question answering and information retrieval tasks.
How It Works
AgriGov employs a schema-driven approach to organize data from government schemes into predefined fields. Automated scraping techniques gather information, which is then translated using a combination of machine translation tools and human editing.
This ensures high-quality, domain-specific translations that are aligned across languages.