Question Answering on MedQA (accuracy, average accuracy)
85.8AccuracyHCQR
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| HCQRLLM Backbone=Qwen3 30B, Context Budget=15 documents2026.03 | 85.8 | — | |
| CoTLLM Backbone=Qwen3 30B, Context Budget=15 documents2026.03 | 84.1 | — | |
| MAIN-RAGLLM Backbone=Qwen3 30B, Context Budget=15 documents2026.03 | 83.6 | — | |
| Rerank-RAGLLM Backbone=Qwen3 30B, Context Budget=15 documents2026.03 | 83.3 | — | |
| RewritingLLM Backbone=Qwen3 30B, Context Budget=15 documents2026.03 | 83.3 | — | |
| HyDELLM Backbone=Qwen3 30B, Context Budget=15 documents2026.03 | 82.9 | — | |
| Simple RAGLLM Backbone=Qwen3 30B, Context Budget=15 documents2026.03 | 82.7 | — | |
| HCQRLLM Backbone=Qwen3 4B, Context Budget=15 documents2026.03 | 77.9 | — | |
| RewritingLLM Backbone=Qwen3 4B, Context Budget=15 documents2026.03 | 73.1 | — | |
| HCQRLLM Backbone=Llama3.1 8B, Context Budget=15 documents2026.03 | 73 | — | |
| CoTLLM Backbone=Qwen3 4B, Context Budget=15 documents2026.03 | 72.8 | — | |
| Rerank-RAGLLM Backbone=Qwen3 4B, Context Budget=15 documents2026.03 | 72.1 | — | |
| HyDELLM Backbone=Qwen3 4B, Context Budget=15 documents2026.03 | 71.4 | — | |
| Simple RAGLLM Backbone=Qwen3 4B, Context Budget=15 documents2026.03 | 71.3 | — | |
| MAIN-RAGLLM Backbone=Qwen3 4B, Context Budget=15 documents2026.03 | 70.9 | — | |
| HyDELLM Backbone=Llama3.1 8B, Context Budget=15 documents2026.03 | 69.3 | — | |
| CoTLLM Backbone=Llama3.1 8B, Context Budget=15 documents2026.03 | 69 | — | |
| RewritingLLM Backbone=Llama3.1 8B, Context Budget=15 documents2026.03 | 67.5 | — | |
| Rerank-RAGLLM Backbone=Llama3.1 8B, Context Budget=15 documents2026.03 | 67.3 | — | |
| Simple RAGLLM Backbone=Llama3.1 8B, Context Budget=15 documents2026.03 | 66.8 | — | |
| MAIN-RAGLLM Backbone=Llama3.1 8B, Context Budget=15 documents2026.03 | 65.8 | — | |
| HCQRLLM Backbone=Llama3.2 3B, Context Budget=15 documents2026.03 | 63.4 | — | |
| HyDELLM Backbone=Llama3.2 3B, Context Budget=15 documents2026.03 | 61 | — | |
| RewritingLLM Backbone=Llama3.2 3B, Context Budget=15 documents2026.03 | 58.9 | — | |
| CoTLLM Backbone=Llama3.2 3B, Context Budget=15 documents2026.03 | 58 | — | |
| Rerank-RAGLLM Backbone=Llama3.2 3B, Context Budget=15 documents2026.03 | 56.8 | — | |
| MAIN-RAGLLM Backbone=Llama3.2 3B, Context Budget=15 documents2026.03 | 55.9 | — | |
| Simple RAGLLM Backbone=Llama3.2 3B, Context Budget=15 documents2026.03 | 55.6 | — | |
| CoTLLM Backbone=Average, Context Budget=15 documents2026.03 | — | 71 | |
| HCQRLLM Backbone=Average, Context Budget=15 documents2026.03 | — | 75 | |
| HyDELLM Backbone=Average, Context Budget=15 documents2026.03 | — | 71.1 | |
| MAIN-RAGLLM Backbone=Average, Context Budget=15 documents2026.03 | — | 69 | |
| Rerank-RAGLLM Backbone=Average, Context Budget=15 documents2026.03 | — | 69.9 | |
| RewritingLLM Backbone=Average, Context Budget=15 documents2026.03 | — | 70.7 | |
| Simple RAGLLM Backbone=Average, Context Budget=15 documents2026.03 | — | 69.1 |