FLANS at SemEval-2026 Task 7: RAG with Open-Sourced Smaller LLMs for Everyday Knowledge Across Diverse Languages and Cultures

Researchers developed FLANS, a system using retrieval-augmented generation with open-source smaller language models for the SemEval-2025 multilingual knowledge task. The system creates culturally-aware knowledge bases from Wikipedia content and integrates live search capabilities, focusing on privacy and sustainability through smaller LLMs deployed on the Ollama platform.

• →FLANS system uses open-source smaller LLMs with retrieval-augmented generation for multilingual question answering tasks.
• →Researchers created culturally-aware knowledge bases by extracting Wikipedia content with prepared keyword lists.
• →The system integrates live online search capabilities through DuckDuckGo for enhanced knowledge retrieval.
• →Focus on privacy and sustainability by deploying smaller LLMs on the Ollama platform rather than large proprietary models.
• →Testing covered English, Spanish, and Chinese languages with shared code and resources available on GitHub.