# Medical Information Retrieval on R2MED (original queries) Source: [Medical Information Retrieval on R2MED (original queries)]() Collection index: [llms.txt]() Data updated: 2026-04-21T11:15:55.962Z Evaluation of Medical Information Retrieval performance using the original query set of the R2MED dataset, measured using various nDCG@10 sub-metrics. 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 Information Retrieval | | Dataset (including any stated split) | R2MED (original queries) | | Metric | Average (nDCG@10) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 19 of 19 | | 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 - [Average (nDCG@10)](): Higher is better; unit: PERCENT. - [Biology (nDCG@10)](): Higher is better; unit: PERCENT. - [Bioin. (nDCG@10)](): Higher is better; unit: PERCENT. - [MedS. (nDCG@10)](): Higher is better; unit: PERCENT. - [MedE. (nDCG@10)](): Higher is better; unit: PERCENT. - [MedD. (nDCG@10)](): Higher is better; unit: PERCENT. - [PMCT. (nDCG@10)](): Higher is better; unit: PERCENT. - [PMCC. (nDCG@10)](): Higher is better; unit: PERCENT. - [IIYiC. (nDCG@10)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Average (nDCG@10) | Result date | Report paper | | --- | --- | --- | --- | | ReasonEmbed-Qwen3-8B | 43.18 | Not recorded | 68fa7c071e209131cea8bfb0 | | ReasonEmbed-Llama-3.1-8B | 42.76 | Not recorded | 68fa7c071e209131cea8bfb0 | | ReasonEmbed-Qwen3-8B | 42.36 | Not recorded | 68fa7c071e209131cea8bfb0 | | ReasonEmbed-Llama-3.1-8B | 41.98 | Not recorded | 68fa7c071e209131cea8bfb0 | | ReasonEmbed-Qwen3-4B | 41.16 | Not recorded | 68fa7c071e209131cea8bfb0 | | ReasonEmbed-Qwen3-4B | 39.94 | Not recorded | 68fa7c071e209131cea8bfb0 | | RaDeR-gte-Qwen2-7B | 35.19 | Not recorded | 68fa7c071e209131cea8bfb0 | | Qwen3-Embedding-8B | 34.22 | Not recorded | 68fa7c071e209131cea8bfb0 | | gte-Qwen2-7B-instruct | 32.56 | Not recorded | 68fa7c071e209131cea8bfb0 | | DIVER-Retriever | 32.23 | Not recorded | 68fa7c071e209131cea8bfb0 | | Qwen3-Embedding-4B | 31.75 | Not recorded | 68fa7c071e209131cea8bfb0 | | NV-Embed-v2 | 31.43 | Not recorded | 68fa7c071e209131cea8bfb0 | | GritLM-7B | 31.12 | Not recorded | 68fa7c071e209131cea8bfb0 | | OpenAI-3-Large | 28.57 | Not recorded | 68fa7c071e209131cea8bfb0 | | ReasonIR-8B | 27.94 | Not recorded | 68fa7c071e209131cea8bfb0 | | Qwen3-8B-ms | 24.15 | Not recorded | 68fa7c071e209131cea8bfb0 | | Qwen3-4B-ms | 23.09 | Not recorded | 68fa7c071e209131cea8bfb0 | | Llama-3.1-8B-ms | 22.34 | Not recorded | 68fa7c071e209131cea8bfb0 | | BM25 | 15.13 | Not recorded | 68fa7c071e209131cea8bfb0 | ### ReasonEmbed-Qwen3-8B Average (nDCG@10): 43.18. Result ID: 69e75a8247f34ce924ce837e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"8B","Training":"Redapter (self-adaptive RI-InfoNCE loss)"}. This record does not assert independent reproduction. ### ReasonEmbed-Llama-3.1-8B Average (nDCG@10): 42.76. Result ID: 69e75a8247f34ce924ce837f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"8B","Training":"Redapter (self-adaptive RI-InfoNCE loss)"}. This record does not assert independent reproduction. ### ReasonEmbed-Qwen3-8B Average (nDCG@10): 42.36. Result ID: 69e75a8247f34ce924ce837b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"8B","Training":"InfoNCE loss"}. This record does not assert independent reproduction. ### ReasonEmbed-Llama-3.1-8B Average (nDCG@10): 41.98. Result ID: 69e75a8247f34ce924ce837c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"8B","Training":"InfoNCE loss"}. This record does not assert independent reproduction. ### ReasonEmbed-Qwen3-4B Average (nDCG@10): 41.16. Result ID: 69e75a8247f34ce924ce837d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"4B","Training":"Redapter (self-adaptive RI-InfoNCE loss)"}. This record does not assert independent reproduction. ### ReasonEmbed-Qwen3-4B Average (nDCG@10): 39.94. Result ID: 69e75a8247f34ce924ce837a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"4B","Training":"InfoNCE loss"}. This record does not assert independent reproduction. ### RaDeR-gte-Qwen2-7B Average (nDCG@10): 35.19. Result ID: 69e75a8247f34ce924ce8378. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"7B"}. This record does not assert independent reproduction. ### Qwen3-Embedding-8B Average (nDCG@10): 34.22. Result ID: 69e75a8247f34ce924ce8373. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"8B"}. This record does not assert independent reproduction. ### gte-Qwen2-7B-instruct Average (nDCG@10): 32.56. Result ID: 69e75a8247f34ce924ce8371. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"7B"}. This record does not assert independent reproduction. ### DIVER-Retriever Average (nDCG@10): 32.23. Result ID: 69e75a8247f34ce924ce8379. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"4B"}. This record does not assert independent reproduction. ### Qwen3-Embedding-4B Average (nDCG@10): 31.75. Result ID: 69e75a8247f34ce924ce8372. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"4B"}. This record does not assert independent reproduction. ### NV-Embed-v2 Average (nDCG@10): 31.43. Result ID: 69e75a8247f34ce924ce8370. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"7B"}. This record does not assert independent reproduction. ### GritLM-7B Average (nDCG@10): 31.12. Result ID: 69e75a8247f34ce924ce836f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"7B"}. This record does not assert independent reproduction. ### OpenAI-3-Large Average (nDCG@10): 28.57. Result ID: 69e75a8247f34ce924ce836e. [Reporting paper]() Recorded evaluation configuration: {"Size":"-"}. This record does not assert independent reproduction. ### ReasonIR-8B Average (nDCG@10): 27.94. Result ID: 69e75a8247f34ce924ce8377. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"8B"}. This record does not assert independent reproduction. ### Qwen3-8B-ms Average (nDCG@10): 24.15. Result ID: 69e75a8247f34ce924ce8375. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"8B","Status":"reproduced"}. This record does not assert independent reproduction. ### Qwen3-4B-ms Average (nDCG@10): 23.09. Result ID: 69e75a8247f34ce924ce8374. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"4B","Status":"reproduced"}. This record does not assert independent reproduction. ### Llama-3.1-8B-ms Average (nDCG@10): 22.34. Result ID: 69e75a8247f34ce924ce8376. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Size":"8B","Status":"reproduced"}. This record does not assert independent reproduction. ### BM25 Average (nDCG@10): 15.13. Result ID: 69e75a8247f34ce924ce836d. [Reporting paper]() Recorded evaluation configuration: {"Size":"-"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.