Terminology Understanding on Medical Domain Dataset
0.979Recall@10LBR
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| LBRBackbone=Qwen2.5-1.5B, Attention Architecture=Causal, Training Paradigm=LLM+IB-GL+CL2026.01 | 0.979 | 0.906 | 87.91 | |
| LBRBackbone=Qwen2-1.5B, Attention Architecture=Causal, Training Paradigm=LLM+IB-GL+CL2026.01 | 0.978 | 0.899 | 87.82 | |
| LBRBackbone=Llama3.2-1B, Attention Architecture=Causal, Training Paradigm=LLM+IB-GL+CL2026.01 | 0.976 | 0.897 | 84.53 | |
| LLM2VecBackbone=SLLaMA-1.3B, Attention Architecture=Bi., Training Paradigm=LLM+CL2026.01 | 0.961 | 0.89 | 79.34 | |
| GTEBackbone=Qwen2-1.5B, Attention Architecture=Bi., Training Paradigm=LLM+CL2026.01 | 0.958 | 0.867 | 65.26 | |
| BGEBackbone=XLMR-Large, Attention Architecture=Bi., Training Paradigm=LLM+CL2026.01 | 0.947 | 0.871 | 78.35 | |
| SFT+CLBackbone=Qwen2-1.5B, Attention Architecture=Causal, Training Paradigm=LLM+GL+CL2026.01 | 0.732 | 0.625 | 64.47 | |
| SFT+CLBackbone=Qwen2.5-1.5B, Attention Architecture=Causal, Training Paradigm=LLM+GL+CL2026.01 | 0.664 | 0.558 | 60.68 | |
| Qwen2Backbone=Qwen2-1.5B, Attention Architecture=Causal, Training Paradigm=LLM+GL2026.01 | 0.41 | 0.335 | 20.01 | |
| Qwen2.5Backbone=Qwen2.5-1.5B, Attention Architecture=Causal, Training Paradigm=LLM+GL2026.01 | 0.353 | 0.286 | 18.311 | |
| HuaTuoBackbone=LLaMA-7B, Attention Architecture=Causal, Training Paradigm=LLM+GL2026.01 | 0.346 | 0.283 | 31.29 | |
| ChemLLMBackbone=InternLM-2B, Attention Architecture=Causal, Training Paradigm=LLM+GL2026.01 | 0.034 | 0.02 | 11.012 |