Update README.md

README.md

@@ -23,7 +23,7 @@ Unlike traditional dense retrievers trained with binary relevance signals, BioHi
 - **Backbone**: BAAI/bge-base-en-v1.5  
 - **Parameters**: ~0.1B  
 - **Fine-tuning**: LoRA (merged into base model)  
-- **Max input length**: 8192 tokens  
+- **Max input length**: 512 tokens  
 - **Training data**: Biomedical abstracts annotated with MeSH labels (e.g., BioASQ-derived corpora)  
 
 ---