Question Answering on MedicalQA
84.2ScoreRSN-Tune
Evaluation Results
| Method | Links | |
|---|---|---|
| RSN-TuneModel=Llama3-8B-it2026.04 | 84.2 | |
| SPAModel=Qwen2.5-14B-it2026.04 | 84.1 | |
| SPAModel=Llama3-8B-it2026.04 | 84 | |
| RandomModel=Llama3-8B-it2026.04 | 83.9 | |
| RandomModel=Qwen2.5-14B-it2026.04 | 83.8 | |
| RSN-TuneModel=Qwen2.5-14B-it2026.04 | 83.5 | |
| SPAModel=Qwen2.5-7B-it2026.04 | 81.7 | |
| RandomModel=Qwen2.5-7B-it2026.04 | 81.4 | |
| RSN-TuneModel=Qwen2.5-7B-it2026.04 | 81 | |
| BaseModel=Llama3-8B-it2026.04 | 80.5 | |
| BaseModel=Qwen2.5-14B-it2026.04 | 80.2 | |
| BaseModel=Qwen2.5-7B-it2026.04 | 77.8 | |
| Non-privateBackbone LLM=GPT-4o2026.06 | 8.81 | |
| SharedRequest (Original)Backbone LLM=GPT-4o2026.06 | 8.74 | |
| Non-privateBackbone LLM=GPT-4o-mini2026.06 | 8.6 | |
| SharedRequest (Original)Backbone LLM=GPT-4o-mini2026.06 | 8.58 | |
| SharedRequest (Simplified)Backbone LLM=GPT-4o2026.06 | 8.4 | |
| SharedRequest (Simplified)Backbone LLM=GPT-4o-mini2026.06 | 8.28 | |
| SharedRequest (Original)Backbone LLM=GPT-3.5-Turbo2026.06 | 6.91 | |
| Non-privateBackbone LLM=GPT-3.5-Turbo2026.06 | 6.89 | |
| SharedRequest (Simplified)Backbone LLM=GPT-3.5-Turbo2026.06 | 6.32 |