Question Answering on MedQA (EM, F1)
71EMAgent
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| AgentBackbone=Qwen3-8B, Budget Strategy=Shared Budget, Avg. Cost=4.38, Avg. Psg=1.13, Avg. Steps=2.132026.06 | 71 | 71 | |
| AgentBackbone=Qwen3-8B, Budget Strategy=Per-query Budget, B = 20, Avg. Cost=5.08, Avg. Psg=1.28, Avg. Steps=2.292026.06 | 68 | 68 | |
| AgentBackbone=Llama3.1-8B-Instruct, Budget Strategy=Per-query Budget, B = 20, Avg. Cost=18.39, Avg. Psg=4.84, Avg. Steps=5.852026.06 | 60 | 60 | |
| AgentBackbone=Llama3.1-8B-Instruct, Budget Strategy=Shared Budget, Avg. Cost=17.84, Avg. Psg=4.72, Avg. Steps=5.742026.06 | 55 | 55 |