Multi-Intent Detection (MID) Model

This model was fine-tuned for the task of Multi-Intent Detection (MID), a type of multi-label classification where each input can have multiple labels assigned. The dataset used for fine-tuning is specifically designed to simplify the MID task, with the number of labels limited to two per instance.

Model Details

Base Model: DeBERTa-v3-base
Task: Multi-label classification
Number of Labels: 2
Fine-tuning Framework: Hugging Face Transformers

Training Configuration

Training Arguments:
- Learning Rate: 2e-5
- Batch Size (Train): 16
- Batch Size (Eval): 16
- Gradient Accumulation Steps: 2
- Number of Epochs: 8
- Weight Decay: 0.01
- Warmup Ratio: 10%
- Learning Rate Scheduler Type: Cosine
- Mixed Precision Training: Enabled (FP16)
- Logging Steps: 50

Performance Metrics

Epoch	Training Loss	Validation Loss	Precision	Recall	F1 Score	Accuracy
0	0.069100	0.069115	0.000000	0.000000	0.000000	0.000000
2	0.024100	0.022929	0.952334	0.316920	0.475576	0.078652
4	0.009200	0.010799	0.959768	0.819894	0.884334	0.653668
6	0.006300	0.008773	0.963243	0.883344	0.921565	0.770654
7	0.006200	0.008707	0.961635	0.886319	0.922442	0.775281

Final Evaluation Metrics (Epoch 8):

Validation Loss: 0.0087
Precision: 0.9616
Recall: 0.8863
F1 Score: 0.9224
Accuracy: 0.7753

Limitations

Simplified Multi-Label Setting: This model assumes a fixed number of two labels per instance, which may not generalize to datasets with more complex multi-label settings.
Performance on Unseen Data: The model's performance may degrade if applied to data distributions significantly different from the training dataset.