WSDM Cup 2026: Multilingual Retrieval Challenge

Participate in WSDM Cup 2026’s multilingual retrieval challenge, a key aspect of RAG systems. Develop systems to search 10M documents in Chinese, Persian, and Russian using English queries.

The WSDM Cup 2026 is calling for participation in a multilingual retrieval task, a crucial aspect of Retrieval-Augmented Generation (RAG) systems. The challenge involves developing systems that can search a collection of 10 million documents in Chinese, Persian, and Russian using English queries.

• Multilingual retrieval task for RAG systems
• Search 10 million documents in Chinese, Persian, and Russian
• English queries used for retrieval
• Submissions must be in TREC run file format

Participants can use 41 development queries and other data like MIRACL for system development. The top teams will receive a slot at WSDM for oral presentation.

For more information and to participate, visit: https://wsdmcup-2026.github.io

Tags: WSDM Cup 2026, Multilingual Retrieval, RAG systems, Information Retrieval, NLP, AI Research