Checklist OPEIA 2025 Competition · Task 2
IOAI Perú: Information Retrieval Challenge
Build and compare TF-IDF and at least one advanced retrieval model over a Spanish corpus about Peru.
The task
Framed around the National Library of Peru's digitisation, the task asks for a Spanish text-normalisation and indexing pipeline, a TF-IDF baseline and at least one advanced model (for example BM25 or embeddings), evaluated with precision, recall and MAP on development queries with relevance judgements, and predictions for test queries.
Abridged by SOTA from the official materials. The official statement has the exact rules, and it wins wherever this summary differs.
At a glance
- You get
- Spanish document corpus, development queries with relevance judgements, test queries (downloaded in the notebook).
- You submit
predictions.txtfor the test queries, the completed notebook, and the best model uploaded to the Hugging Face Hub.- Scoring
- Standard IR metrics (precision, recall, MAP); the exact scoring formula is not stated in the notebook.
- Format
- OPEIA 2025, a 14-day team competition (teams of four listed on the results page); dates not stated on the pages opened.