multilingual-retrieval

Code for our winning submission to CheckThat! 2026 Task 1: Source Retrieval for Scientific Web Claims. Given a scientific claim from social media, the task is to retrieve the referenced scientific paper from a large candidate collection. Our approach ranked 1st out of 37 participating teams. 🏆

A reproducible benchmark for incomplete-cue personal mobility recall, comparing BM25, multilingual E5 dense retrieval, and dense-seeded graph retrieval with RRF/PPR diagnostics.

Multilingual sentence embeddings for 10 Northeast Indian languages