Multi-domain Language Modeling on MGSM, HumanEval, and MedMCQA composite (test)
46.1Average Score (Composite)LoRAFlow
Evaluation results
| Method | Links | |
|---|---|---|
| LoRAFlowBase LLM=LLaMA-3-8B, Paradigm=Supervised2025.12 | 46.1 | |
| qa-FLoRABase LLM=LLaMA-3-8B, Paradigm=Data & Training free2025.12 | 44.2 | |
| Avg [0.5, 0.5]Base LLM=LLaMA-3-8B, Paradigm=Static fusion2025.12 | 38.5 | |
| Centroid sim.Base LLM=LLaMA-3-8B, Paradigm=Training free2025.12 | 34.4 | |
| LoRAFlowBase LLM=LLaMA-2-7B, Paradigm=Supervised2025.12 | 30.9 | |
| LoRAHubBase LLM=LLaMA-2-7B, Paradigm=Supervised2025.12 | 26.4 | |
| qa-FLoRABase LLM=LLaMA-2-7B, Paradigm=Data & Training free2025.12 | 25.8 | |
| Avg [0.5, 0.5]Base LLM=LLaMA-2-7B, Paradigm=Static fusion2025.12 | 20.4 | |
| Centroid sim.Base LLM=LLaMA-2-7B, Paradigm=Training free2025.12 | 18.8 |