@inproceedings{almanza-etal-2026-verbanexai,
    title = "{V}erba{N}ex{AI} at {S}em{E}val-2026 Task 7: Integrating Web Snippets and {RAG} for the Evaluation of Multilingual Cultural Knowledge in {LLM}s",
    author = "Almanza, Danileth  and
      Serrano, Jairo  and
      Puertas, Edwin  and
      Martinez Santos, Juan Carlos",
    editor = "Kochmar, Ekaterina  and
      Ghosh, Debanjan  and
      North, Kai  and
      Komachi, Mamoru",
    booktitle = "Proceedings of the 20th {I}nternational {W}orkshop on {S}emantic {E}valuation (2026)",
    month = jul,
    year = "2026",
    address = "San Diego, California, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl-workshops/2026.semeval-1.123/",
    pages = "899--904",
    ISBN = "979-8-89176-414-9",
    abstract = "In multilingual and multicultural contexts, LLMs require contextualization mechanisms to generate culturally coherent responses. In this sense, this study presents a LLaMA-based approach to answer short cultural questions in different languages within Task 7 of SemEval-2026 (Track 1: SAQ), without access to official training data. The system integrates controlled synthetic data generation, evidence retrieval through web snippets, and a Retrieval-Augmented Generation (RAG) framework with Few-shot learning. BLEnD is used solely as a thematic guide, ensuring semantic independence. During development, the LLaMA-3.1-8B model achieved 38.51{\textbackslash}{\%} global accuracy, while LLaMA-3.2-1B obtained 15.54{\textbackslash}{\%}. In large-scale evaluation (30,500 instances), the 1B model achieved 16.69{\textbackslash}{\%}, maintaining stability after prompt optimization. The results demonstrate that contextual retrieval improves multilingual cultural knowledge evaluation and highlight the importance of pipeline design and model capacity."
}