Rating Prediction on Amazon Musical Instruments 5-core (test)
0.5271MAEREVLoRA (Llama 1B)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| REVLoRA (Llama 1B)Evaluation Protocol=Fine-Tuned, Model Backbone=Llama 3.2 1B, Optimization Loss=MAE2025.12 | 0.5271 | 0.9233 | |
| Llama 3BEvaluation Protocol=Few-Shot, Shot-count=3-shot, Model Backbone=Llama 3.2 3B2025.12 | 0.5312 | 1.0556 | |
| LURPEvaluation Protocol=Few-Shot, Shot-count=3-shot, Rating-only=true2025.12 | 0.5411 | 1.0556 | |
| REVLoRA (Qwen 0.5B)Evaluation Protocol=Fine-Tuned, Model Backbone=Qwen 2.5 0.5B, Optimization Loss=MAE2025.12 | 0.5473 | 1.028 | |
| Qwen 0.5BEvaluation Protocol=Few-Shot, Shot-count=3-shot, Model Backbone=Qwen 2.5 0.5B2025.12 | 0.556 | 1.0514 | |
| Llama 3BEvaluation Protocol=Zero-Shot, Model Backbone=Llama 3.2 3B2025.12 | 0.5575 | 1.0877 | |
| Qwen 3BEvaluation Protocol=Few-Shot, Shot-count=3-shot, Model Backbone=Qwen 2.5 3B2025.12 | 0.5609 | 1.0792 | |
| Qwen 3BEvaluation Protocol=Zero-Shot, Model Backbone=Qwen 2.5 3B2025.12 | 0.5676 | 0.9343 | |
| Llama 1BEvaluation Protocol=Zero-Shot, Model Backbone=Llama 3.2 1B2025.12 | 0.5858 | 1.1389 | |
| DIRECTEvaluation Protocol=Deep Learning2025.12 | 0.6263 | 0.7252 | |
| Qwen 0.5BEvaluation Protocol=Zero-Shot, Model Backbone=Qwen 2.5 0.5B2025.12 | 0.633 | 0.9374 | |
| RGCLEvaluation Protocol=Deep Learning2025.12 | 0.6443 | 0.7704 | |
| REVLoRA (Llama 1B MSE)Evaluation Protocol=Fine-Tuned, Model Backbone=Llama 3.2 1B, Optimization Loss=MSE2025.12 | 0.6596 | 0.7377 | |
| Llama 1BEvaluation Protocol=Few-Shot, Shot-count=3-shot, Model Backbone=Llama 3.2 1B2025.12 | 0.6735 | 1.4903 | |
| DeepCoNNEvaluation Protocol=Deep Learning2025.12 | 1.1885 | 1.8998 |