Update README.md

README.md

@@ -1,18 +1,21 @@
+---
+license: mit
+base_model:
+- utter-project/EuroLLM-9B-Instruct
+---
 
-# EuroLLM-Classifier-QLoRA
+# EuroLLM  QLoRA – Grounding Act Classification
 
-This repository contains a German binary sentence classifier fine-tuned on top of **EuroLLM** using **QLoRA**.
+This model is a fine-tuned version of utter-project/EuroLLM-9B-Instruct optimized using QLoRA for efficient binary classification of German dialogue utterances into:
 
-## Task
+ADVANCE: Contribution that moves the dialogue forward (e.g. confirmations, follow-ups, elaborations)
 
-The model classifies utterances into two categories:
+NON-ADVANCE: Other utterances (e.g. vague responses, misunderstandings, irrelevant comments)
 
-- `ADVANCE`: Contributions that move the conversation forward (e.g., confirmations, suggestions, answers).
-- `NON_ADVANCE`: Utterances that do not advance the dialogue (e.g., misunderstandings, clarifications, hesitation).
-
-## Base Model
-
-The base model used is: `utter-project/EuroLLM-9B-Instruct`
+## Use Cases
+Dialogue system analysis
+Teacher-student interaction classification
+Grounding in institutional advising or classroom discourse
 
 ## How to Use
 
@@ -23,4 +26,4 @@ from peft import PeftModel, PeftConfig
 peft_config = PeftConfig.from_pretrained("MB55/EuroLLM-Classifier-QLoRA")
 base_model = AutoModelForSequenceClassification.from_pretrained(peft_config.base_model_name_or_path)
 model = PeftModel.from_pretrained(base_model, "MB55/EuroLLM-Classifier-QLoRA")
-tokenizer = AutoTokenizer.from_pretrained("MB55/EuroLLM-Classifier-QLoRA")
+tokenizer = AutoTokenizer.from_pretrained("MB55/EuroLLM-Classifier-QLoRA")