Multilingual Text Classification With Relevance Scoring

via “cross-lingual document reranking with relevance scoring”

Cohere's reranking model boosting search relevance 20-40%.

Unique: Uses cross-attention mechanism to jointly encode query-document pairs rather than separate embeddings, enabling fine-grained relevance assessment across 100+ languages without language-specific model variants. Achieves 20-40% precision improvement when inserted into existing retrieval pipelines (BM25, vector, hybrid) without requiring retriever retraining.

vs others: Outperforms embedding-based reranking (which uses separate query/document encodings) by capturing query-document interaction patterns; faster to integrate than retraining retrievers and language-agnostic unlike monolingual ranking models.

via “multilingual-text-classification-with-relevance-scoring”

text-classification model by undefined. 98,81,128 downloads.

Unique: 3-way classification head (relevant/irrelevant/neutral) trained on 2.7B query-passage pairs with hard negative mining, enabling nuanced relevance filtering beyond binary classification; XLM-RoBERTa backbone provides zero-shot multilingual transfer without language-specific fine-tuning

vs others: More granular than binary relevance classifiers (includes neutral class for ambiguous cases) and more efficient than ensemble approaches; single model handles 100+ languages vs maintaining separate classifiers per language

via “multilingual relevance scoring with xlm-roberta backbone”

text-classification model by undefined. 31,06,509 downloads.

Unique: Leverages XLM-RoBERTa's 100-language pretraining with BAAI's domain-specific fine-tuning on English-Chinese relevance pairs, enabling zero-shot cross-lingual scoring without separate language models or translation pipelines

vs others: Simpler and faster than translation-based reranking (query translation + monolingual scoring) while achieving comparable accuracy, and more cost-effective than proprietary multilingual APIs