Update README.md

README.md

@@ -18,7 +18,7 @@ metrics:
 - type: loss
   value: 0.2673
 model-index:
-- name: PubMedBERT NLI LoRA
+- name: PubMedBERT BioNLI LoRA
   results:
   - task:
       type: natural-language-inference
@@ -35,14 +35,14 @@ model-index:
       value: 0.2673
 ---
 
-# PubMedBERT NLI (LoRA Fine-Tuned)
+# PubMedBERT BioNLI LoRA
 
-[![Model](https://img.shields.io/badge/Model-LoRA-green)](https://huggingface.co/Bam3752/Model-lora)
+[![Model](https://img.shields.io/badge/Model-LoRA-green)](https://huggingface.co/Bam3752/PubMedBERT-BioNLI-LoRA)
 [![Hugging Face](https://img.shields.io/badge/HF-Bam3752-blue)](https://huggingface.co/Bam3752)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
 
-A biomedical **Natural Language Inference (NLI)** model based on **PubMedBERT** and fine-tuned with **LoRA** adapters.  
-It is designed to evaluate **entailment, contradiction, and neutrality** between reasoning steps in biomedical chain-of-thought (CoT) explanations.
+**PubMedBERT BioNLI LoRA** is a biomedical **Natural Language Inference (NLI)** model fine-tuned with **LoRA adapters**.  
+It classifies **entailment, contradiction, and neutrality** between biomedical text pairs, optimized for **chain-of-thought reasoning validation**.
 
 ---
 
@@ -50,8 +50,8 @@ It is designed to evaluate **entailment, contradiction, and neutrality** between
 
 - **Base model:** [pritamdeka/PubMedBERT-MNLI-MedNLI](https://huggingface.co/pritamdeka/PubMedBERT-MNLI-MedNLI)  
 - **Fine-tuning datasets:** BioASQ + MedNLI  
-- **Objective:** 3-class NLI (entailment, neutral, contradiction)  
-- **Method:** LoRA (PEFT) adapters  
+- **Objective:** 3-class NLI (entailment / neutral / contradiction)  
+- **Method:** LoRA parameter-efficient fine-tuning  
 - **Hardware:** Apple MPS (Metal backend)  
 - **Hyperparameters:**  
   - Epochs: 4  
@@ -72,19 +72,19 @@ It is designed to evaluate **entailment, contradiction, and neutrality** between
 | Macro F1    | **0.9036** |
 | Eval Loss   | **0.2673** |
 
-Calibrated with isotonic regression (`calibration/isotonic.pkl`) for more reliable probability estimates.
+👉 Calibrated with isotonic regression (`calibration/isotonic.pkl`) for reliable probabilities.
 
 ---
 
 ## 🚀 Usage
 
-### Load with Transformers
+### Transformers
 
 ```python
 from transformers import AutoModelForSequenceClassification, AutoTokenizer
 
-model = AutoModelForSequenceClassification.from_pretrained("Bam3752/Model-lora")
-tokenizer = AutoTokenizer.from_pretrained("Bam3752/Model-lora")
+model = AutoModelForSequenceClassification.from_pretrained("Bam3752/PubMedBERT-BioNLI-LoRA")
+tokenizer = AutoTokenizer.from_pretrained("Bam3752/PubMedBERT-BioNLI-LoRA")
 
 premise = "Aspirin reduces the risk of myocardial infarction."
 hypothesis = "Aspirin prevents heart attacks."