# Medical Question Answering on MedQA (Decision-Useful Rate) Source: [Medical Question Answering on MedQA (Decision-Useful Rate)]() Collection index: [llms.txt]() Data updated: 2026-03-20T10:19:07.069Z Evaluation of model performance on the MedQA dataset for Medical Question Answering, reporting the Decision-Useful Rate. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Medical Question Answering | | Dataset (including any stated split) | MedQA | | Metric | Decision-Useful Rate | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 30 of 30 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Decision-Useful Rate](): Higher is better; unit: PERCENT. ## Recorded results | Method | Decision-Useful Rate | Result date | Report paper | | --- | --- | --- | --- | | HCQR | 89.8 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | HCQR | 83.2 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | HCQR | 82.1 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | HCQR | 81.8 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | HCQR | 73.4 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | Rewriting | 65.6 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | Rewriting | 62.2 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | Rewriting | 56.1 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | HyDE | 54.6 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | HyDE | 54.5 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | HyDE | 53.4 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | HyDE | 52.8 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | HyDE | 51.6 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | Rewriting | 51.2 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | Rewriting | 45.2 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | Rerank-RAG | 44.1 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | Rerank-RAG | 44.1 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | Rerank-RAG | 44.1 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | Rerank-RAG | 44.1 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | Rerank-RAG | 44.1 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | Simple RAG | 30 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | Simple RAG | 30 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | Simple RAG | 30 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | Simple RAG | 30 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | Simple RAG | 30 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | MAIN-RAG | 26.7 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | MAIN-RAG | 26.4 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | MAIN-RAG | 25.4 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | MAIN-RAG | 24.7 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | | MAIN-RAG | 23.7 | Not recorded | 69bcd4aa0636e5bf4afc2be0 | ### HCQR Decision-Useful Rate: 89.8. Result ID: 69bd1eff47f34ce924a7de52. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3 30B"}. This record does not assert independent reproduction. ### HCQR Decision-Useful Rate: 83.2. Result ID: 69bd1eff47f34ce924a7de51. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3 4B"}. This record does not assert independent reproduction. ### HCQR Decision-Useful Rate: 82.1. Result ID: 69bd1eff47f34ce924a7de53. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Average"}. This record does not assert independent reproduction. ### HCQR Decision-Useful Rate: 81.8. Result ID: 69bd1eff47f34ce924a7de50. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama3.1 8B"}. This record does not assert independent reproduction. ### HCQR Decision-Useful Rate: 73.4. Result ID: 69bd1eff47f34ce924a7de4f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama3.2 3B"}. This record does not assert independent reproduction. ### Rewriting Decision-Useful Rate: 65.6. Result ID: 69bd1eff47f34ce924a7de43. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3 30B"}. This record does not assert independent reproduction. ### Rewriting Decision-Useful Rate: 62.2. Result ID: 69bd1eff47f34ce924a7de42. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3 4B"}. This record does not assert independent reproduction. ### Rewriting Decision-Useful Rate: 56.1. Result ID: 69bd1eff47f34ce924a7de44. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Average"}. This record does not assert independent reproduction. ### HyDE Decision-Useful Rate: 54.6. Result ID: 69bd1eff47f34ce924a7de48. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen3 30B"}. This record does not assert independent reproduction. ### HyDE Decision-Useful Rate: 54.5. Result ID: 69bd1eff47f34ce924a7de46. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Llama3.1 8B"}. This record does not assert independent reproduction. ### HyDE Decision-Useful Rate: 53.4. Result ID: 69bd1eff47f34ce924a7de49. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Average"}. This record does not assert independent reproduction. ### HyDE Decision-Useful Rate: 52.8. Result ID: 69bd1eff47f34ce924a7de47. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen3 4B"}. This record does not assert independent reproduction. ### HyDE Decision-Useful Rate: 51.6. Result ID: 69bd1eff47f34ce924a7de45. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Llama3.2 3B"}. This record does not assert independent reproduction. ### Rewriting Decision-Useful Rate: 51.2. Result ID: 69bd1eff47f34ce924a7de41. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama3.1 8B"}. This record does not assert independent reproduction. ### Rewriting Decision-Useful Rate: 45.2. Result ID: 69bd1eff47f34ce924a7de40. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama3.2 3B"}. This record does not assert independent reproduction. ### Rerank-RAG Decision-Useful Rate: 44.1. Result ID: 69bd1eff47f34ce924a7de3b. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama3.2 3B"}. This record does not assert independent reproduction. ### Rerank-RAG Decision-Useful Rate: 44.1. Result ID: 69bd1eff47f34ce924a7de3c. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama3.1 8B"}. This record does not assert independent reproduction. ### Rerank-RAG Decision-Useful Rate: 44.1. Result ID: 69bd1eff47f34ce924a7de3d. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3 4B"}. This record does not assert independent reproduction. ### Rerank-RAG Decision-Useful Rate: 44.1. Result ID: 69bd1eff47f34ce924a7de3e. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3 30B"}. This record does not assert independent reproduction. ### Rerank-RAG Decision-Useful Rate: 44.1. Result ID: 69bd1eff47f34ce924a7de3f. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Average"}. This record does not assert independent reproduction. ### Simple RAG Decision-Useful Rate: 30. Result ID: 69bd1eff47f34ce924a7de36. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Llama3.2 3B"}. This record does not assert independent reproduction. ### Simple RAG Decision-Useful Rate: 30. Result ID: 69bd1eff47f34ce924a7de37. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Llama3.1 8B"}. This record does not assert independent reproduction. ### Simple RAG Decision-Useful Rate: 30. Result ID: 69bd1eff47f34ce924a7de38. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen3 4B"}. This record does not assert independent reproduction. ### Simple RAG Decision-Useful Rate: 30. Result ID: 69bd1eff47f34ce924a7de39. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen3 30B"}. This record does not assert independent reproduction. ### Simple RAG Decision-Useful Rate: 30. Result ID: 69bd1eff47f34ce924a7de3a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Average"}. This record does not assert independent reproduction. ### MAIN-RAG Decision-Useful Rate: 26.7. Result ID: 69bd1eff47f34ce924a7de4d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3 30B"}. This record does not assert independent reproduction. ### MAIN-RAG Decision-Useful Rate: 26.4. Result ID: 69bd1eff47f34ce924a7de4b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama3.1 8B"}. This record does not assert independent reproduction. ### MAIN-RAG Decision-Useful Rate: 25.4. Result ID: 69bd1eff47f34ce924a7de4e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Average"}. This record does not assert independent reproduction. ### MAIN-RAG Decision-Useful Rate: 24.7. Result ID: 69bd1eff47f34ce924a7de4c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3 4B"}. This record does not assert independent reproduction. ### MAIN-RAG Decision-Useful Rate: 23.7. Result ID: 69bd1eff47f34ce924a7de4a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama3.2 3B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.