Update README.md

Updated README with updated model trained from mmBERT

README.md

@@ -61,6 +61,8 @@ Improves general linguistic coverage.
 
 ## Pretraining Strategy
 
+![paper_diagram](https://cdn-uploads.huggingface.co/production/uploads/690e53164670480464c0d207/JcNksj4nBefZ87WtxWOTl.png)
+
 We adopt a **two-stage domain-adaptive pretraining approach**:
 
 ### Stage 1: Mixed-domain pretraining
@@ -78,25 +80,27 @@ We adopt a **two-stage domain-adaptive pretraining approach**:
 
 ## Evaluation
 
-We evaluate MultiClinicalBERT on **11 clinical NLP tasks across 5 languages**:
+![main_results](https://cdn-uploads.huggingface.co/production/uploads/690e53164670480464c0d207/in7j8ifSGRl35mL2KrLlK.png)
+
+We evaluate MultiClinicalBERT on **10 clinical NLP tasks across 5 languages**:
 
-- English: MIMIC-III Mortality, MedNLI, MIMIC-IV CDM  
+- English: MIMIC-III Mortality, MIMIC-IV CDM  
 - Chinese: CEMR, IMCS-V2 NER  
-- Japanese: IFMIR NER, IFMIR Incident Type  
-- Russian: RuMedNLI, RuCCoNNER  
-- Spanish: De-identification, PPTS  
+- Japanese: IFMIR Incident Type, IFMIR NER
+- Russian: RuMedNLI, RuCCoN NER
+- Spanish: EHR De-identification, PPTS  
 
 ### Key Results
-- Consistently outperforms multilingual BERT (mBERT)
+- Consistently outperforms Modern multilingual BERT (mmBERT)
 - Matches or exceeds strong language-specific models
 - Largest gains observed in low-resource settings
-- Statistically significant improvements (Welch’s t-test, p < 0.05)
 
 Example:
-- MedNLI: **83.90% accuracy**
-- CEMR: **86.38% accuracy**
-- IFMIR NER: **85.53 F1**
-- RuMedNLI: **78.31% accuracy**
+- MIMIC-III Mortality: **89.40% accuracy**
+- MIMIC-IV CDM: **92.18% accuracy**
+- CEMR: **94.24% accuracy**
+- IFMIR NER: **86.21 F1**
+- EHR De-identification: **88.64 F1**
 
 
 ## Key Contributions