andreaschari/bge-m3-ZH_MMARCO_NATIVE

BGE-m3 ZH mMARCO/v2 Native Queries

This is a BGE-M3 model post-trained on the Chinese dataset from MMARCO/v2.

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

Model Details

Model Description

• —Model Type: Sentence Transformer <!-- - Base model: Unknown -->
• —Maximum Sequence Length: 8192 tokens
• —Output Dimensionality: 1024 tokens
• —Similarity Function: Cosine Similarity <!-- - Training Dataset: Unknown --> <!-- - Language: Unknown --> <!-- - License: Unknown -->

Training Details

Framework Versions

• —Python: 3.10.13
• —Sentence Transformers: 3.1.1
• —Transformers: 4.45.1
• —PyTorch: 2.4.1
• —Accelerate: 0.34.2
• —Datasets: 3.0.1
• —Tokenizers: 0.20.3