Update README.md

README.md

@@ -20,27 +20,45 @@ This is a fine-tuned **Language-Agnostic BERT Sentence Embedding (LaBSE)** model
 * **Base Architecture:** `sentence-transformers/LaBSE`
 * **Task:** Text Classification (Intent Recognition)
 * **Number of Classes:** 77 (Derived from the BANKING77 taxonomy)
-* **Supported Languages:** English, Sinhala, Tamil, Singlish (Code-mixed), and Tanglish (Code-mixed).
 
-## Use Case
-This model is intended to be used as the **Digital Text Router** in a hybrid multimodal pipeline. When a user submits a typed text ticket (without an image/screenshot), this model deeply analyzes the semantic context of the text, regardless of the language, to instantly categorize the ticket for the correct banking department.
+## Supported Languages & Scripts
+This model has been robustly fine-tuned to handle both native scripts and Romanized (code-mixed) phonetic typing, which is highly prevalent in South Asian digital communication.
 
-## Performance & Ablation Findings
-During our extensive ablation studies, this Transformer model was evaluated against traditional Machine Learning algorithms (like Linear SVMs). 
+| Language | Script | Example Input |
+|---|---|---|
+| **English** | Latin | *"I lost my credit card yesterday"* |
+| **Sinhala** | Sinhala | *"මගේ කාඩ් එක නැතිවුනා"* |
+| **Singlish** | Latin (Code-mixed) | *"mage card eka nathi wela"* |
+| **Tamil** | Tamil | *"என் கார்டு தொலைந்துவிட்டது"* |
+| **Tanglish** | Latin (Code-mixed) | *"en card tholainthu vittathu"* |
 
-* **Digital/Clean Text:** This LaBSE model achieves state-of-the-art semantic understanding on clean, typed digital text.
-* **OCR Noisy Text:** When exposed to high Character Error Rates (CER) from Tesseract OCR (e.g., extracting text from blurry mobile screenshots), this model's F1-score drops to ~49.70% because Transformers are highly sensitive to sub-word tokenization destruction. 
-* **Architecture Decision:** Because of this sensitivity, the Swift backend dynamically routes **OCR-extracted text** to a robust Linear SVM, while routing **clean digital text** to this LaBSE model.
+## Use Case & Architecture
+This model is intended to be used as the **Digital Text Router** in a hybrid multimodal pipeline. When a user submits a typed text ticket (without an image/screenshot), this model deeply analyzes the semantic context of the text to instantly categorize the ticket for the correct banking department.
 
-## How to use in Python
+## Evaluation & OCR Ablation Results
+During our ablation studies, this Transformer model was evaluated against traditional Machine Learning algorithms (Linear SVMs) to determine its robustness to OCR-induced Character Error Rates (CER). 
 
-You can easily use this model via the `transformers` pipeline:
+The F1-scores below represent how well the models retain intent-classification accuracy when forced to read noisy OCR text (Tesseract) extracted from varying image qualities, compared to clean ground-truth text.
+
+| Image Quality Condition | LaBSE Raw OCR vs Clean | LaBSE+SpellCheck vs Clean | SVM Raw OCR vs Clean |
+|---|---|---|---|
+| `clean` | 49.70% | 31.17% | **94.15%** |
+| `blur` | 39.56% | 30.74% | **45.32%** |
+| `rotation` | 48.29% | 26.58% | **69.30%** |
+| `low-resolution` | 30.05% | 25.47% | **30.36%** |
+| **OVERALL** | 34.18% | 23.42% | **35.37%** |
+
+**Conclusion:** 
+Transformers like LaBSE are highly sensitive to sub-word tokenization destruction caused by OCR typos. Because of this architectural sensitivity, the Swift backend dynamically routes **OCR-extracted text** to a robust Linear SVM (which proved to be 94.15% resilient on clean OCR), while routing **clean typed digital text** to this LaBSE model to leverage its superior deep semantic understanding.
+
+## How to use via Serverless API
 
 ```python
-from transformers import pipeline
+import requests
 
-classifier = pipeline("text-classification", model="Swift-Support/labse-intent-1.0")
+API_URL = "https://api-inference.huggingface.co/models/Swift-Support/labse-intent-1.0"
+headers = {"Authorization": "Bearer YOUR_HF_TOKEN"}
 
-result = classifier("I lost my credit card yesterday, please help me cancel it")
-print(result)
-# Output: [{'label': 'Card payment declined', 'score': 0.98}]
+# Sending a Singlish code-mixed ticket
+response = requests.post(API_URL, headers=headers, json={"inputs": "mage card eka nathi wela"})
+print(response.json())