New training support makes custom token-level search models easy

Original: Training and Finetuning Multi-Vector Embedding Models with Sentence Transformers

Source: huggingface.co ↗