Update README.md

README.md

@@ -24,9 +24,9 @@ Unlike response-generation models, this version is trained to **predict client f
 
 ## 💡 Overview
 
-- ✅ Fine-tuned on **2,601 dialogues** with client feedback scores between **70 and 98**
-- ✅ Data collected through **text-based role-play** by trained counselors 
-- ✅ Covers a wide range of topics: depression, family, school, career, relationships, and more
+- ✅ Task: Predict the **overall counseling quality score** as rated by the client
+- ✅ Dataset: 6,589 dialogues with feedback scores between 0 and 100
+- ✅ Data source: Text-based role-play by trained counselors
 - ✅ Base Model: [`tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.3`](https://huggingface.co/tokyotech-llm/Llama-3.1-Swallow-8B-Instruct-v0.3)
 
 ---
@@ -95,17 +95,15 @@ Fine-tuning was performed using **QLoRA** with the following configuration:
 
 ### Dataset Split
 
-- **Training Data**: 2,601 dialogues with feedback scores between 70 and 98  
-- **Train/Validation Split**: 90% train, 10% validation
+- **Training/Validation/Test ratio**: 8:1:1
 
 ### Hyperparameter Settings
 
-- **Optimizer**: `adamw_8bit`
+- **Optimizer**: `adamw_torch_fused`
 - **Warm-up Steps**: `100`
-- **Learning Rate**: `1e-3`
-- **Epochs**: `5`
-- **Batch Size**: `8`
-- **Validation Frequency**: every 400 steps
+- **Learning Rate**: `2e-4`
+- **Epochs**: `4`
+- **Batch Size**: `4`
 
 ---
 
@@ -131,5 +129,6 @@ If you use this model or dataset, please cite the following paper:
   - [KokoroChat on GitHub (UEC-InabaLab)](https://github.com/UEC-InabaLab/KokoroChat)
 - 🤖 **Model Variants**:  
   - [Llama-3.1-KokoroChat-Low](https://huggingface.co/UEC-InabaLab/Llama-3.1-KokoroChat-Low): fine-tuned on **3,870 dialogues** with client feedback scores **< 70**
+  - [Llama-3.1-KokoroChat-High](https://huggingface.co/UEC-InabaLab/Llama-3.1-KokoroChat-High): fine-tuned on **2,601 dialogues** with client feedback scores between **70 and 98**
   - [Llama-3.1-KokoroChat-Full](https://huggingface.co/UEC-InabaLab/Llama-3.1-KokoroChat-Full): fine-tuned on **6,471 dialogues** with client feedback scores **≤ 98**
 - 📄 **Paper**: [ACL 2025 Paper](https://aclanthology.org/2025.acl-long.608/)