---
license: mit
datasets:
- unicamp-dl/mmarco
language:
- zh
base_model:
- unicamp-dl/mt5-base-mmarco-v2
---

# mt5-base Reranker ZH mMARCO/v2 Transliterated Queries tokenised with Anserini

This is a variation of Unicamp's [mt5-base Reranker](https://huggingface.co/unicamp-dl/mt5-base-mmarco-v2) initially finetuned on mMARCOv/2. 

The queries are transliterated from Chinese to English text using [uroman](https://github.com/isi-nlp/uroman).
The queries were tokenised with [pyterrier_anserini](https://github.com/seanmacavaney/pyterrier-anserini/tree/main/pyterrier_anserini).

The model was used for the SIGIR 2025 Short paper: Lost in Transliteration: Bridging the Script Gap in Neural IR.